DDX3X syndrome: a multicenter genotype-phenotype study

DDX3X syndrome: a multicenter genotype-phenotype study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article DDX3X syndrome: a multicenter genotype-phenotype study Emanuele G. Coci, Christoph G.W. Gertzen, Pilar Chacon Millan, and 64 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7842722/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract DDX3X dysfunction causes an X-linked multisystem disorder with high penetrance and variable expressivity. The phenotypic spectrum spans from learning disability without somatic involvement to profound intellectual disability with severe impairments in the central nervous system and other organs. A few multicenter studies and single case reports have previously highlighted some common phenotypic patterns but were unable to delineate correlations between underlying DDX3X variants and phenotypic findings. From the second largest patient cohort published to date, we analysed clinical, psychometric and diagnostic findings of 52 female and 7 male individuals, harbouring de novo and inherited DDX3X variants. These female patients revealed previously unknown quantitative correlations between variant type and localization and specific phenotypic findings (growth features, epilepsy, brain anomalies, dysmorphisms, motor-focused neurological findings). Moreover, by analysing the in silico folding and RNA binding capability of mutant DDX3X monomers, we were able to delineate novel correlations between DDX3X monomer misfolding grade and phenotypic severity. DDX3X associated multisystem disorder DDX3X variants’ type and position DDX3X genotype-phenotype correlation In-silico simulation of mutatnt DDX3X monomers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Key Message Our DDX3X patient cohort with 59 individuals is the second largest published ever We found correlations between variants’ type and position and phenotypical features We simulated mutant DDX3X monomers via AlphaFold2 We compared mutant DDX3X monomers’ folding and RNA binding to wild-type DDX3X monomer Introduction The clinical management of genetically determined neurodevelopmental disorders is a major effort for medical institutions, even more when the clinical presentation of a disorder is highly variable among affected individuals. DEAD-box helicase 3 ( DDX3X ) is member of DExD/H-box RNA helicase superfamily, maps in Xp11.4 band and encodes for an RNA-binding protein involved in entire RNA homeostasis [ 1 , 2 ]. The helicase activity is performed by two enzymatic domains localized in the central part of the monomer, and the overall function is performed by a DDX3X homodimer. DDX3X syndrome was initially described in 2015 [ 3 ]. Since then, DDX3X variants have been reported in a few studies [ 4 – 6 ] and in several case reports [ 7 – 10 ]; clinical heterogeneity is clearly recognizable in DDX3X deficiency: cognitive impairment spans from learning disability to profound intellectual disability and organic dysfunction is heterogeneous among patients [ 11 – 15 ]. In female cases, DDX3X escapes X chromosome inactivation, and this phenomenon complicates the delineation of genotype-phenotype correlations [ 16 ]. In the case of heterozygous missense and in-frame deletion variants, some female patients’ cells simultaneously produce wild-type, mixed and mutant DDX3X homodimers in a total physiologic quantity; whereas heterozygous frameshift, stop-codon and splice site variants in female patients’ cells exclusively produce wild-type DDX3X homodimers in reduced quantity. Male patients harbor exclusively missense variants, whereas male embryos with loss of function variants die in utero ; with hemizygous missense variants, male patients’ cells exclusively produce mutant DDX3X homodimers in physiologic amounts. Although genetic, clinical and diagnostic data have been broadly reported, no study has focused on quantitative correlation between DDX3X variant type and localization and clinical, psychometric and diagnostic findings. Based on 52 female and 7 male patients, we have studied the second largest DDX3X patient cohort to date and have searched for quantitative correlations between underlying DDX3X variant type and localization and phenotypic findings. As a first for DDX3X neurodevelopmental syndrome, we performed molecular simulations for 24 DDX3X monomers, carrying single amino acid variations from 18 female and 6 male included missense variants, and classified the variations as to their clinical severity according to their folding free energy and RNA binding ability. Materials and Methods Patients We identified and accrued female and male individuals with neurodevelopmental phenotype who harbor a DDX3X variant using GeneMatcher [ 17 ] and through a search of the Baylor Genetics clinical database. We excluded individuals who carried pathogenic or likely pathogenic variants in other human disease genes. Our cohort consisted of fifty-two female and seven male patients whose medical data had been previously collected and recorded at the managing clinical institution during regular clinical interactions. DDX3X sequencing DDX3X variants were identified in the probands using massively parallel sequencing (next generation sequencing) based technologies (exome/genome sequencing with or without employing virtual gene panels) in either clinical diagnostic or research settings. Parental testing was performed in most cases, either as part of trio exome/genome sequencing or by testing for the specific variant by Sanger sequencing. Data analysis, variant filtering, and prioritization were performed using the in-house implemented pipelines of the local genetic centers. Pathogenicity of the identified DDX3X variants was established according to American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) criteria (Richards et al., 2015). All the variants are described based on the NM_001356.5 (GRCh37/hg19) transcript of DDX3X in according to Human Genome Variation Society recommendations ( https://varnomen.hgvs.org/ ). All variants were confirmed using Mutalyzer ( https://mutalyzer.nl ). Statistical analysis Continuous and categorical variables for genotype-phenotype correlations were analyzed by performing Pearson Chi-square test in R version 4.3.2. DDX3X monomer modelling Modeling and Molecular Dynamics simulations The wild-type structure of DDX3X in an RNA-bound state was generated by deleting one copy of DDX3X from the X-ray crystal structure (PDB-ID: 6O5F) [ 18 ]. Subsequently, gaps within the structure were filled using Prime and Maestro ( Schrödinger, LLC, New York, USA , 2017). The variations Ser181Thr, Phe182Val, Ile191Asn, Gln207Glu, Leu220Ser, Gly227Arg, Leu235Gln, Thr275Met, Arg326Cys, Arg326His, Arg351Gln, Arg376Cys, Ala404Asp, Gly406Arg, Trp421Gly, Leu427Gln, Leu484Pro, Arg488Cys, and Thr498Arg were created in Maestro, and for all variations and the wild-type the protonation states were assigned using PROPKA at pH 7.4 [ 19 ]. Molecular dynamics (MD) simulations were performed with Amber22 [ 20 ]. The ff14SB force field was used to parameterize the protein, Joung-Chetham parameters were used for the counter ions, and TIP3P for the water. Here, the ff99 force field [ 21 ] with the “OL3” χ distribution from ff14SB [ 22 , 23 ] and the parmbsc0 α/γ [ 24 ] modifications resulting in the Amber ff99OL3 parameters were used for the RNA. For a description of how we generated input structures and thermalized the simulation systems, please see “Supplementary Methods”. Twelve independent production runs of MD simulations with a constant number of particles, constant volume, and constant temperature (NVT) with 1 µs length each were performed. For this, the starting temperatures of the MD simulations at the beginning of the thermalization were varied by a fraction of a Kelvin. Unbinding of RNA was evaluated using the Virtual Molecular Dynamics (VMD) program [ 25 ]. To assess the potentially detrimental effects of the variations on folding, we generated an AlphaFold2 [ 26 ] model of DDX3X to gain information about structural regions with variations not resolved in the X-ray crystal structure. This model was subsequently analyzed with FoldX [ 27 ]. First, van der Waals clashes in the wild-type DDX3X were removed via the “optimize” command. Then, all residues in the protein were varied to all other residues and itself to judge the effect of the variation on folding. A repetition of this procedure revealed no differences between the computed energies for all variations. Classification of severity based on folding free energy and MD simulations of RNA binding To classify the variations into those with a more severe or less severe clinical outcome, their folding free energy and their capability to have RNA remaining bound in MD simulations were used as descriptors. A perceptron was trained using a Python3.9 script; see “Supplementary Methods” for “Details for training a perceptron” as to the software libraries used. The script linearly scales the data from 0 to 1, where 1 means a better clinical outcome for the patient or a better-predicted function of DDX3X. Where no MD data was available, a function similar to the wild-type was assumed, as other strategies common in data preparation for machine learning, e.g., using the mean of the data set, resulted in worse predictions. Such variations are found in the unstructured termini of DDX3X, thus, they are unlikely to affect RNA binding. This is corroborated by no predictive models being found when assuming otherwise. The dataset was split into female and male patients, and for each group, the computational predictions were correlated with the severity score for each of four clinical outcomes (neurological dysfunction, epilepsy, brain anomalies, and dysmorphisms) via a perceptron using the “minimize" function of SciPy. Here, “Nelder-Mead” was used as a method with 10,000 maximum iterations to fit the decision boundary for the perceptron. The severity score is transformed into a binary classifier during this step according to a cutoff, which is varied from 0 to 0.9 in steps of 0.1. Due to sparse training data (no. of data points per clinical outcome < 7 for male and < 19 for female patients), no test set was used. All values equal to or below the cutoff are classified as class 0 (more severe) and all values above the cutoff are classified as class 1 (less severe). A cutoff of 1 would result in all severities classified as class 0 such that no predictions were possible. To train the perceptron, a linear decision boundary was fitted by minimizing a function that changes the slope and y-axis intercept of the decision boundary toward an optimal F1 score. See “Supplementary Methods” for “Details for line fitting”. Results DDX3X variants Our cohort includes 52 female (88%) and 7 male (12%) patients, including 1 pair of female siblings and 1 pair of male siblings. Out of 52 female patients, 47 harbor de novo variants, and 5 could not be tested for inheritance due to absent parental DNA samples. Out of 7 male patients, 2 harbor de novo variants and 5 maternally transmitted variants. Each patient harbors one single variant in DDX3X , heterozygous in female and hemizygous in male individuals. In 52 female patients, we report 46 different variants: 20 missense (43.5%), 5 in-frame deletion (10.9%), 7 stop-gain (15.2%), 7 splice site (15.2%), 6 frameshift deletion (13.0%) and 1 frameshift duplication (2.2%); in 7 male patients, we report 6 different missense variants (Fig. 1 A). Seven variants (3 missense, 2 stop-gain, 1 in-frame deletion, 1 frameshift) are reported twice. Out of 59 patients, 35 patients (28 female and 7 male) display 31 different amino acid variations, which are caused by 26 missense variants and 5 in-frame deletion variants; four amino acid variations are reported twice (Fig. 1 B). Twenty-two out of 52 female patients harbor a missense variant and 2 missense variants were identified twice, and all 7 male patients harbor missense variants ( p < 0.05 ) (Fig. 2 A). Out of 20 different missense variants (43.5%) from female patients, 18 (90%) cause a variation among different amino acid classes and 2 (10%) a variation within the same amino acid class ( p < 0.001 ) (Fig. 2 B). Out of 6 missense variants revealed in male patients, 3 (50%) cause a variation among different amino acid classes and 3 (50%) a variation within the same amino acid class. Pooling together 20 missense and 5 in-frame deletion variants from female patients, 18 (72%) map in exons coding for helicase domains and their binding tract, which span 55.2% of monomer length, and 7 (28%) map in exons coding N-terminal regions, which span 44.8% of monomer length ( p < 0.01 ) (Fig. 2 C). Out of 6 missense variants from male patients, 2 (33%) affect amino acids in a helicase domain, and 4 (66%) affect amino acids in the N-terminal or C-terminal regions (Fig. 2 D). Twenty-four female patients harbor 14 loss of function (nonsense, stop-gain, frameshift deletion, frameshift duplication) variants and 7 splice site variants (together 45.7% of 46 different female variants). None of the 21 nonsense and splice site variants and none of the 25 missense and in-frame deletion variants from our female cohort map in 3` terminal DDX3X portion, which encodes the C-terminal region (13.1% of monomer length) (Fig. 2 E). Clinical findings and correlation to underlying DDX3X variants We have analyzed the entire spectrum of clinical and diagnostic findings from 52 female and 7 male patients according to six categories (growth features, intellectual delay, epilepsy, brain anomalies, dysmorphisms, neurological findings); separately for female and male patients, we have searched for quantitative correlations between these phenotypic findings and underlying DDX3X variant types and, for missense variants, class and localization of the mutated amino acid. The entire clinical and diagnostic data set for female patients is reported in Supplementary Table 1 (Suppl. Table 1) and for male patients in Supplementary Table 2 (Suppl. Table 2). Somatometric features Birth somatometric parameters (weight, length, and head circumference) are fully reported for 38 out of 52 female patients; incomplete natal parameters are reported for 10 female patients and no natal parameters are reported for 4 female patients. Looking for a quantitative correlation between low prenatal growth and underlying DDX3X variants, we have focused on patients with natal parameters below or equal to 10th centile. Thirteen patients with natal weight below or equal to 10th centile are reported, harboring 2 in-frame deletion variants (15.4%), 7 missense variants (53.8%) and 4 nonsense/splice site variants (30.8%); twelve patients with natal length below or equal to 10th centile are reported, harboring 7 missense variants (58.3%) and 5 nonsense/splice site variants (41.7%); twelve patients with natal head circumference below or equal to 10th centile are reported, harboring 3 in-frame deletion variants (25%), 5 missense variants (41.7%) and 4 nonsense/splice site variants (33.3%). Measurements for all three somatometric parameters at the last follow-up are reported for 46 out of 52 female patients. In 5 female patients, only two parameters are reported and in 1 female patient no parameter is reported. We have searched for a quantitative correlation between somatometric parameters below or equal to the 10th centile and underlying DDX3X variants. A weight below or equal to 10th centile is reported in 15 patients, harboring 1 an in-frame deletion variant (6.7%), 11 a missense variant (73.3%) and 3 a nonsense/splice site variant (20%); a significant correlation is found between missense and in-frame deletion variants and low weight (p < 0.01) and between missense variants and low weight ( p < 0.05 ). A length below or equal to the 10th centile is reported in 15 patients, harboring 3 in-frame deletion variants (20%), 10 missense variants (66.7%) and 2 nonsense/splice site variants (13.3%); a significant correlation is found between missense and in-frame deletion variant types and low length (p < 0.001) . A head circumference below or equal to the 10th centile is reported in 16 patients, harboring 1 in-frame deletion variant (6.3%), 9 missense variants (56.3%) and 6 nonsense/splice site variants (37.5%) (Fig. 3 A and 3 B). In female patients harboring a missense or in-frame deletion variant and displaying length below or equal to 10th centile at birth and at last follow-up, the variant is localized in helicases coding tracts more frequently than in the N-terminal region coding tract ( p < 0.01 for birth length; p < 0.05 for last registered length). In female patients harboring a missense or in-frame deletion variant and displaying weight below or equal to 10th centile at birth, the variant is localized in helicases coding tracts more frequently than in the N-terminal region coding tract ( p < 0.05 ). In female patients harboring a missense or in-frame deletion variant and displaying a head circumference below or equal to the 10th centile at birth and at the last follow-up, the variant is localized in the N-terminal region coding tract as prevalently as in helicases coding tracts (Fig. 3 C). Out of 7 male patients harboring missense variants, natal somatometric parameters are fully reported in 2, partially reported in 2, and not reported in 3 patients. Natal weight is reported in 4 patients, and none has a value below or equal to the 10th centile. Natal length is reported in 3 patients and none has a value below or equal to the 10th centile. Natal head circumference is reported in 3 patients and 1 patient has a value below 10th centile (33.3%). In 6 out of 7 male patients, somatometric parameters are also registered at the last follow-up. Weight is registered in 4 patients and is below 10th centile in 2 of them (50%). Length is registered in 4 patients and was below 10th centile in 2 of them (50%). Head circumference is registered in 5 patients and is below 10th centile in 1 of them (20%). In male patients displaying somatometric parameters below or equal to the 10th centile, a significant prevalence of localization of three missense variants (one in C-terminal helicase and two in non-enzymatic regions) in a specific DDX3X tract cannot be revealed due to a small sample size. Cranial, facial, skeletal, and somatic dysmorphisms We have quantified the dysmorphologic grade, searching for structural anomalies in ten body parts (skull, eyes, ears, philtrum/lips, palate, nose, chin, skeletal parts including hands and feet, internal organs, skin) and scoring each part with 1 point, if it displays at least one anomaly. Using a 10-point dysmorphology score, we have defined subgroup 1 (score from 0 to 3), subgroup 2 (score from 4 to 6) and subgroup 3 (score from 7 to 10). Dysmorphic findings are reported in 51 out of 52 female patients. Using the 10-point dysmorphology score, we have divided 51 female patients into subgroup 1 (31.4%), subgroup 2 (51.1%), and subgroup 3 (17.6%), revealing the different distribution of patients in the 3 subgroups to be statistically significant ( p < 0.01 ) (Fig. 4 A). Among 16 patients in subgroup 1, 5 patients (31.2%) harbor nonsense variants, 8 patients (50%) missense variants, and 3 patients (18.7%) in-frame deletion variants. Among 26 patients in subgroup 2, 15 patients (57.7%) harbor nonsense variants, 10 patients (38.5%) missense variants, and 1 patient (3.8%) in-frame deletion variant. Among 9 patients in subgroup 3, 4 patients (44%) harbor nonsense variants, 4 patients (44%) missense variants, and 1 patient (12%) in-frame deletion variant. A direct correlation between dysmorphology score and variant types is not revealed; nevertheless, a direct correlation between dysmorphology score and localization of amino acid variations in helicase domains is revealed among patients with missense variants ( p < 0.05 ) (Fig. 4 B). Dysmorphic findings are reported in all 7 male patients. Using the 10-point dysmorphology score, we have divided them into subgroup 1 (71.4%) and subgroup 2 (28.6%), and a correlation between dysmorphology score and missense variant localization is not evident, likely due to small sample size. Neurological dysfunction Neurological dysfunction score addresses muscle tone, muscle reflex, gait, balance and cranial nerves and was reported in 48 out of 52 female patients. We have divided the 48 female patients into four subgroups according to neurological dysfunction grade: 6 patients with normal motor development for age (score 0), 10 patients with isolated hypotonia (score 1), 13 patients with hypotonia plus one further neurological finding (score 2), and 19 patients with hypotonia plus two or more neurological findings (score 3). Missense and in-frame deletion variants are seen in 15 out of 19 patients with score 3 (78.9%) and in 14 out of 29 patients with scores 0, 1, and 2 (48.3%); splice site/stop-codon/frame-shift variants are seen in 4 out of 19 patients with score 3 (21.1%) and in 15 out of 29 patients with score 0, 1 and 2 (51.7%). In female patients with neurological dysfunction grade 3, missense and in-frame deletion variants were revealed more frequently than nonsense and splice site variants ( p < 0.01 ) (Fig. 5 A). Out of 25 female patients presenting neurological dysfunction of any grade and harboring missense and in-frame deletion variants, 6 patients (24%) have variants in N-terminal region coding tract and 19 patients (76%) in helicases coding tracts ( p < 0.001 ) (Fig. 5 B); all 3 female patients (100%) without neurological dysfunction have 2 missense variants in N-terminal region coding tract. Neurological findings were reported in all 7 male patients; 4 patients were reported without any neurological dysfunction, 1 with a neurological dysfunction score 1 and 2 with a neurological dysfunction score 3. In the 3 male patients with neurological dysfunction, two missense variants map in N-terminal and C-terminal coding regions and do not lead to amino acid class change; one missense variant (Arg488Cys) maps in a helicase-coding region and leads to an amino acid class change. Epilepsy and EEG findings Out of 52 female patients, 10 have displayed epileptic seizures (19.2%) in the form of absences (6 patients), generalized seizures (4 patients), focal and complex-focal seizures (3 patients), and Blitz-Nick-Salaam (BNS) seizures (1 patient). In 10 female patients with epileptic seizures, missense variants (6 patients, 60%) and in-frame deletion variants (2 patients, 20%) are together more prevalent than splice site variants (2 patients, 20%) ( p < 0.05 ). In 42 female patients without epileptic seizures, missense variants are seen in 16 patients (38.1%), in-frame deletion variants in 4 patients (9.5%), frameshift variants in 8 patients (19%), stop-gain variants in 9 patients (21.4%) and splice site variants in 5 patients (11.9%) (Fig. 6 A). In 8 female patients displaying epilepsy and harboring missense or in-frame deletion variants, 2 patients (25%) have the variant in the N-terminal region coding tract and 6 patients (75%) in gene tracts coding helicase domains and binding tract (Fig. 6 B). Out of 7 male patients, 2 patients have displayed epileptic generalized seizures (28.6%), and both associated amino acid variations (Arg603Gln and Gly607Ala) lie in the C-terminal region; since variation Gly607Ala is seen in one twin with and one twin without seizures, a correlation between epileptic seizures and C-terminal localization of this amino acid variation cannot be done. Brain anomalies Out of 52 female patients, 51 underwent brain MRI scans and are subdivided into two groups according to normal or pathologic findings of brain MRI. The prevalence of variant types (nonsense/splice site variants versus missense/in-frame deletion variants) within these 51 female patients is analyzed; twenty-three patients have nonsense or splice site variants, 7 (30.4%) displaying pathological brain MRIs and 16 (69.6%) normal brain MRIs ( p < 0.05 ), and twenty-eight patients have missense and in-frame deletion variants, 20 (71.4%) displaying pathological brain MRIs and 8 (28.6%) normal brain MRIs ( p < 0.01 ) (Fig. 7 A). Among 24 female patients with normal brain MRI scans, 8 patients (33.3%) have missense and in-frame deletion variants and 16 patients (66.6%) have nonsense and splice site variants ( p < 0.05 ); out of 8 female patients with normal brain MRI scans and missense or in-frame deletion variants, 6 patients (75%) have the variant in N-terminal region coding tract and 2 patients (25%) in helicase domains coding tracts. Among 27 female patients with pathologic brain MRI findings, 20 patients (74.1%) have missense or in-frame deletion variants and 7 patients (25.9%) have nonsense and splice site variants ( p < 0.01 ); out of 20 female patients with pathologic brain MRI findings and missense or in-frame deletion variants, 3 patients (15%) harbor the variant in N-terminal region coding tract and 17 patients (85%) in tracts coding helicase domains and binding tract ( p < 0.001 ) (Fig. 7 B and 7 C). Among 27 female patients with pathological MRI findings, we distinguish 9 patients (subgroup 1) with isolated ventricular enlargement or isolated corpus callosum anomalies and 18 patients (subgroup 2) with multiple anomalies in two or more different brain regions. Within subgroup 1, 4 patients (44.4%) harbor nonsense variants and 5 patients (55.6%) missense and in-frame deletion variants; within subgroup 2, 3 patients (16.7%) harbor nonsense variants and 15 patients (83.3%) missense and in-frame deletion variants ( p < 0.001 ) (Fig. 7 D). Among 20 female patients displaying pathological brain MRI findings and harboring missense and in-frame deletion variants, 15 patients (75%) are included within subgroup 2 and 5 patients (25%) within subgroup 1 ( p < 0.01 ) (Fig. 7 E). Among 7 male patients, 6 patients underwent brain MRI studies; no brain anomalies are reported in 3 patients and pathological MRI findings are reported in 3 patients. In the 3 patients without brain anomalies in MRI scans, both underlying missense variants map within the C-terminal region; in the 3 patients with pathological MRI findings, 2 missense variants map in the N-terminal region and 1 missense variant in the ATP-binding helicase domain. Intellectual disability Out of 52 female patients, IQ is measured in 20 patients by using one of the different standard psychometric tests and intellectual disability (ID) grade is diagnosed on measured IQ; IQ test is not performed in the remaining 24 patients and ID grade is estimated on several clinical observations. In 8 further patients, an ID grade cannot be reported by the managing clinician. We have gathered all 44 female patients with measured (20) ID as well as clinically estimated (24) ID, grouped them into three ID subgroups (mild-to-moderate ID, moderate-to-severe ID, severe ID) and looked for quantitative distribution of variant subsets within the 3 subgroups. Missense variants of female patients are more prevalent in the severe ID subgroup (50%) than in the other two ID subgroups (15% in mild-to-moderate ID subgroup and 35% in moderate-to-severe ID subgroup), being this different prevalence not statistically significant. The entire distribution of the variant subsets in the psychometric subgroups from measured and clinically estimated female patients is reported in “Supplementary Results”. Out of 7 male patients, the ID grade could be estimated in 5 patients; among these 5 patients, IQ could be measured in 4 patients. No association can be revealed between ID grade and variations´ position or between ID grade and variations´ class change. Molecular modeling and simulations of DDX3X variations Female variations are predicted to be more detrimental to DDX3X folding than male variations To test our hypothesis that variations found in female patients are more detrimental to protein function than variations found in male patients, we used a structural model created via AlphaFold2 [ 26 ] to predict the effect on the free energy of folding via FoldX [ 27 ]; positive values indicate that the variation is less stable than the wild-type. The variations found in males Gly117Val, Ser181Thr, Arg351Gln, Arg488Cys, Arg603Gln, and Gly607Ala and the variations found in females Pro40Leu, Pro41His, Phe182Val, Ile191Asn, Gln207Glu, Leu220Ser, Gly227Arg, Leu235Gln, Thr275Met, Arg326Cys, Arg326His, Arg376Cys, Ala404Asp, Gly406Arg, Trp421Gly, Lys427Gln, Leu484Pro, and Thr498Arg were analyzed. The highest detrimental effect on the DDX3X folding free energy was 2.34 kcal mol − 1 (R603Q) among variations from males and 6.44 kcal mol − 1 (Gly227Arg; Fig. 8 A) among variations from females with a mean folding free energy over all variations from males of 0.63 kcal mol − 1 and over all variations from females of 2.55 kcal mol − 1 . Five out of the six variations found in male patients led to a folding free energy change lower than 1.4 kcal mol − 1 , whereas the same was predicted for only three out of 19 variations found in female patients. The cutoff of 1.4 kcal mol − 1 was used as this signifies a ten-times lower likelihood of being correctly folded compared to the wild-type. Comparing the detrimental energies of the male and female cohorts with a Mann-Whitney U test [ 28 , 29 ] reveals a statistically significant difference ( p < 0.002 ) between the two, which shows that the variations found in female patients are significantly more detrimental to protein folding than the variations found in males. The male variations are mainly (five of six) located on the protein surface, whereas many of the female variations appear in the hydrophobic protein core (eight of 18) (Fig. 8 B). This is also true for the most detrimental female variation Gly227Arg, which increases the size of the sidechain by six heavy atoms; several variations from females are small-to-large changes in the sidechain. Yet, also variations with a negative (i.e., favorable compared to the wild-type) change in the folding free energy were identified. Still, these variations might impact the RNA binding capability or the helicase activity of DDX3X. MD simulations suggest reduced RNA binding in DDX3X variations As some variations might lead to reduced RNA binding even if properly folded, we performed MD simulations of the DDX3X wild-type and the Ser181Thr, Phe182Val, Ile191Asn, Gln207Glu, Leu220Ser, Gly227Arg, Leu235Gln, Thr275Met, Arg326Cys, Arg326His, Arg351Gln, Arg376Cys, Ala404Asp, Gly406Arg, Trp421Gly, Leu427Gln, Leu484Pro, Arg488Cys, and Thr498Arg variations in the presence of a bound RNA double strand. Not all variations analyzed in the previous section could be included in the MD analysis, as their locations were not resolved in the X-ray crystal structure with bound RNA. The X-ray crystal structure of DDX3X co-crystallized with RNA was used (PDB ID 605F 1 ) to generate starting structures for the wild-type and the variations. In the MD simulations, the RNA remained bound to the DDX3X wild-type in seven out of twelve replicas. Variations in the DDX3X monomer from all patients ( p = 0.04 ), from female patients ( p = 0.04 ), with folding free energy < 1.4 kcal mol − 1 ( p = 0.04 ), with negative folding free energy ( p = 0.06 ), and from male patients ( p = 0.07 ) confer a reduced RNA binding property compared to DDX3X wild-type (Fig. 8 C). Thus, MD simulations show that RNA is less likely to stay bound in mutated DDX3X than in DDX3X wild-type; this reduction is statistically significant in the first three subgroups. Predictions of folding free energy and RNA binding can be used to classify variation outcomes into more severe and less severe clinical cases We hypothesized that variations found in female patients have a more detrimental effect on protein folding and function than those found in male patients, and our computational predictions correlate well with this hypothesis. We thus probed next whether the predicted effects also correlate with the severity of the assessed clinical outcomes. To do so, we trained perceptron models with a linear decision boundary towards an optimal F1-score. Initially, we encoded the clinical outcomes reported above into a linearized severity score ranging from 0 to 1. For neurological dysfunction, a severity score of 0 (best outcome) corresponds to a clinical score of 0 (normal motor development) and a severity score of 1 (worst outcome) corresponds to a clinical score of 3 (hypotonia plus two or more neurological findings). Choosing a cutoff within the severity score range provides a binary classification where patients with scores above and including the cutoff are treated as “more severe” and those with scores below the cutoff are treated as “less severe”. This allows us to assign different weights to the categories “more severe” and “less severe”. For the above mentioned example of neurological dysfunction, a cutoff of 0.7 would only classify variants with a clinical score of 2 to 3 as “more severe”. For four clinical outcomes (neurological dysfunction, epilepsy, brain anomalies, dysmorphisms) in female as well as male patients, a perceptron model was trained using the predicted protein folding and function data as input to best categorize the patients. The goodness of the categorization is assessed by F1 and ROC-AUC scores; F1 score ranges from 0 (no classification) to 1 (perfect classification) and includes both precision and recall, ROC-AUC score ranges from 0 (inverse perfect classification) through 0.5 (random classification) to 1 (perfect classification). The perceptron models yielded F1 scores up to 1.0 (Fig. 9 A) and ROC-AUC scores up to 1.0 (Fig. 9 B). Generally, models with the highest F1 scores are based on severity score cutoffs between 0.7 and 1. Yet, no satisfying classifications for the outcomes “brain anomalies” and “dysmorphisms” in male patients could be identified for the cutoffs 0.5 to 1; for lower cutoffs, models with F1-scores ~ 0.8 are found. Exemplary categorization results into “more severe” and “less severe” cases, based on the combination of protein function and folding properties, are shown together with the decision boundary determined by the perceptron (Fig. 9 C and 9 D). The model for neurological dysfunction in female patients shows that the decision boundary (green dotted line) is close to the line of ideal anticorrelation with respect to protein function and folding; it means that less severe cases are characterized by good protein function and folding and vice versa (Fig. 9 C). The model for epilepsy in female patients (Fig. 9 D) shows a decision boundary with a lower slope than in the model for neurological dysfunction (Fig. 9 C), indicating that the protein folding properties are less determining for classification. Discussion DDX3X syndrome is one of the most prevalent genetic neurodevelopmental disorders. Defining robust correlations between underlying DDX3X variants and phenotypic findings would sharpen the clinical prognostic view after genetic diagnosis and might facilitate the entire clinical management. using a cohort of 52 female and 7 male patients, our study analyses the second-largest DDX3X patient cohort published so far. By considering type and localization of identified DDX3X variants and the clinical and diagnostic findings of affected patients, we have delineated novel features caused by DDX3X variants and have revealed previously unidentified correlations between DDX3X genotypes and specific pathophenotypes. Variant types and localization and class of variations differ among sexes The subset of missense and in-frame deletion variants as well as the subset of stop-gain, frameshift and splice site variants are revealed in similar proportions within our female cohort, showing that a quantitatively normal population of wild-type, mixed and mutant DDX3X homodimers as well as a reduced population of wild-type DDX3X homodimers are both pathogenic. While no variants map in the final two exons (16 and 17) in our female patients, a few stop-gain variants (Tyr576*, Arg602*, Arg603*) and one mosaic missense variant (Arg602Gln) were reported in those two exons in previously reported female patients [ 4 , 5 ]. It might imply that dysfunction of one single C-terminal region is acceptable in mixed DDX3X homodimer and that almost all female individuals harboring DDX3X missense variants in exons 16 and 17 are asymptomatic and remain undiagnosed. Our study includes the largest male DDX3X cohort analyzed to date. Five novel amino acid variations (Gly117Val, Ser181Thr, Arg488Cys, Arg603Gln, Gly607Ala) enlarge the current view on male DDX3X genetics [ 3 – 6 , 9 , 30 ]. Out of six male missense variants, two are de novo and four are inherited from healthy mothers, implying that the effect of these missense variants of hemizygous male patients is too weak in heterozygous females to cause a manifest disease. In line with the paradigmatic healthiness of variant-transmitting heterozygous mothers from our and previous DDX3X studies, missense variants of our male patients were not revealed in our female patients. Nevertheless, our de novo male variant Arg488Cys was reported in a previous female patient [ 5 ] and this is, to our knowledge, the first overlap of one DDX3X variant among a male and a female patient; based on this first observation, we firstly question the above-mentioned paradigm. While no variation has been reported in C-terminal region in a male before, our two novel variations Arg603Gln and Gly607Ala map distal to C-terminal helicase domain and this shed new light on function of C-terminal region in mutant DDX3X homodimer. Regarding the localization of variations, missense variants from our patients differ among sexes; female variations are more prevalent in helicase domains (14) than in N-terminal region (5) and, contrarily, male variations are more prevalent in N- and C-terminal regions (4) than in helicase domains (2). Merging 11 male variations (9 in helicase domains and 2 in N-terminal region) from former studies with our 5 novel male variations, we note a higher prevalence in helicase domains (11 variations) than in N- and C-terminal regions (6 variations) like for female patients, although the ratio of variations in enzymatic tracts to variations in not-enzymatic tracts is less marked in males than in females. Looking at the class of variations in our female and male patients, they clearly differ among sexes; in female patients, variations into a different amino acid class are 9 times more frequent than those into the same amino acid class, probably because variations of a wild-type amino acid into a similar one does not impair the function of mixed DDX3X homodimer strongly enough to cause a clinical pathophenotype. This female discrepancy in amino acid class change is not observed in our male patients, who display amino acid variations into the same class as frequently as into another class, and it likely indicates that each mutated DDX3X homodimer from a hemizygous variant is pathogenic. Uniting 10 former male variations (8 within the same class and 2 into another class) together with our 5 novel male variations, we note a higher prevalence of change into another amino acid class (11) than within the same one (5) like for female patients, although the male ratio is lower than in female patients. Novel correlations in our female patients and an association trend in our male patients emerge between DDX3X genotype and specific pathophenotypes In our female cohort, missense and in-frame deletion variants are more prevalently associated with low weight and length than nonsense and splice site variants; this gap increases from natal age to last follow-up age and might indicate that mutant and mixed DDX3X homodimers have a more detrimental impact on corporal growth than decreased wild-type DDX3X homodimers. Somatometric centiles decrease also in males over time. This phenomenon, firstly described in our study, might imply that DDX3X has a stronger role in cell proliferation during postnatal growth than during fetal period. A dysmorphology score for DDX3X syndrome was first established in our study. We have observed that a direct correlation between dysmorphology score and localization of missense variants in helicase coding regions characterizes female patients; in male patients, no correlation between dysmorphology score and localization of missense variants has been identified. We first implemented a motor-focused neurological score in DDX3X clinical research and correlated it to underlying DDX3X zygosity. Two interesting neurological observations have risen up in our female cohort: first, the more severe the neurological dysfunction, the more frequent a missense or in-frame deletion variant; second, females with a missense variant in helicase-coding regions always display a neurological dysfunction, and this is relevant for clinical management and genetic consulting. Since neurological findings were not or were partially described in former male patients, previous data cannot be used to depict any correlation between male neurological dysfunction grade and variations’ features, and our male patients’ data are unconclusive in this regard. Epilepsy is much more prevalent in DDX3X patients (about 20%) than in the general population, considering together female and male patients from our study. Since our female epileptic patients harbor missense and in-frame deletion variants more prevalently than nonsense variants and since missense variants are more frequent in helicase coding regions than in N-terminal coding regions in this patient subgroup, it might imply that mutant and mixed DDX3X homodimers cause more severe electrophysiological neuronal dysfunction than quantitatively reduced wild-type DDX3X homodimers. As for neurological findings, our data on male epilepsy do not lead to any correlative conclusion. Two correlations between variant type and brain MRI findings and, respectively, localization of missense and in-frame deletion variants and brain MRI findings have emerged in our female cohort. Grouping female patients with pathologic MRI findings in two severity subgroups (1 and 2), we have observed that severe findings (subgroup 2) are more prevalently associated to missense and in-frame deletion variants than to nonsense variants, as observed in a former study focusing on polymicrogyria [ 5 ]; it implies that mutated and mixed DDX3X homodimers cause more severe structural brain anomalies than quantitatively reduced wild-type DDX3X homodimers. A clear correlation between localization of missense and in-frame deletion variants in helicase domains and pathologic brain MRI has risen up as well. Although a trend between variant localization in N-terminal region and pathologic brain findings can be observed in our male patients, the low number of brain MRIs does not allow any statistical conclusion, and future male DDX3X patients will show whether proximal position of missense variants is or isn’t associated to structural brain anomalies. Intellectual disability is a well-defined phenotypic feature of DDX3X syndrome. In our female patients, missense variants were observed more prevalently in the severe ID subgroup (50%) than in the moderate-to-severe (35%) and mild-to-moderate (15%) ID subgroups. Although this trend is biased by an unclear delimitation of ID subgroups’ borders and has to be analyzed more precisely in future DDX3X studies, it might imply that mutant and mixed DDX3X homodimers cause a more severe neuronal dysfunction than reduced wild-type DDX3X homodimers and therefore give rise to severe ID grade. In our male patients, no correlation between genotype features and phenotype findings has been noted. Taken together the entire female data, missense and in-frame variant types as well as localization of amino acid variations in helicase domains are directly correlated to a severe grade in five of six analyzed pathophenotypic findings. Considering the entire male data, one association trend rises up between missense variant localization and brain anomalies. Molecular simulations identify sex-specific effects in DDX3X variations The three-times higher prevalence of DDX3X variations in female patients compared to male patients suggests that more variant-harbouring male embryos die in utero than females. Hence, we hypothesized that variations in females cause DDX3X to be non-functional whereas variations in males only reduce the functionality of DDX3X. We used FoldX to predict how severe the impact of a variation on folding is compared to wild-type [ 31 ], as a misfolded protein will not be functional. In the FoldX analysis, variation-bearing monomers in females had on average a 1.92 kcal mol − 1 higher detrimental energy, i.e., they are on average 25 times less likely to correctly fold than variation-bearing monomers in males. The average variation in males is only three times less likely to be correctly folded compared to wild-type. A variation from a female leading to strong misfolding likely leads to death when occurring in a male embryo and, therefore, we have not found any overlap between our female and male variations. Nevertheless, the first overlap of a variation (Arg488Cys, 0.79 kcal mol − 1 ) between our male patient and a former female patient might raise the hypothesis that a variation can cause a clinical pathophenotype in both sexes if the associated folding free energy is not too high, allowing the viability of a male embryo, and not too low, causing clinically relevant findings in a female individual. According to this hypothesis, male variation Arg603Gln (2.34 kcal mol − 1 ) should be revealed also in future female patients and its report in our current male patient might be clarified through its localization in a not-enzymatic region. In female patients with a wild-type copy, the phenotype should be less drastic as at least a quarter of all DDX3X dimers should be folded correctly. Most of the variations from males and some variations from females resulted in monomer folding free energies < 1.4 kcal mol − 1 implying that their folding behavior is not much different from the DDX3X wild-type. Thus, we hypothesized that these variations should reduce the functionality of DDX3X, e.g., by decreasing the RNA binding capability. To probe this, we performed MD simulations of one DDX3X monomer (see below) in its folded state in the presence of RNA. We included all variations that could be mapped to an X-ray crystal structure of DDX3X in a complex with RNA (PDB ID: 6O5F) [ 18 ] and found that DDX3X variations are significantly less likely to stay bound to RNA compared to the wild-type. This holds true for variations with a FoldX energy < 1.4 kcal mol − 1 , which are likely to fold correctly; hence, these variations likely hamper the functionality of DDX3X by lowering its affinity towards RNA rather than leading to a misfolded protein. We used a monomeric form of DDX3X bound to RNA in the MD simulations as monomers bind to dsRNA such that each protomer recognizes one RNA strand, i.e., the RNA binding is independent of a dimer state [ 18 ]. The dimers then cooperate in the unwinding of the dsRNA [ 18 ]. Hence, this setup allowed us to study the RNA binding capabilities of DDX3X variations in a time-efficient manner. We did not study deletion variations in our computational approaches as the exact effect of the deletion on the structure is difficult to predict. Finally, several variations, e.g., Pro40Leu, Pro41His, Arg603Gln, and Gly604Ala, are located in the intrinsically disordered regions (IDRs) of the C- and N-terminus of DDX3X; as IDRs often convey interactions between proteins [ 32 , 33 ] and DDX3X has several non-conventional functions facilitated by interaction with other proteins [ 34 ], the variations may influence these non-canonical functions. Finally, we identified two models to classify patients into more severe and less severe outcomes via machine learning. In some models, the predicted variation effect on protein folding has a lower impact on severity than the predicted effect on protein function. This indicates that overall the correct folding of DDX3X might influence fetal survival, where we do not see severe outcomes in this case, which might apply to male patients in our study, whereas the functional impairment could be more important for the clinical outcome, which might apply to male and female patients in our study. Due to the sparse data used for training the machine learning model, the model predictions should not guide health-related behavior and/or clinical decisions. Declarations Data availability All data are within the paper and its Supporting Information files. Acknowledgments We thank the parents and caregivers of all included patients as well as the medical and para-medical personnel involved in patients’ medical care. Author contributions The study was concepted byEmanuele G. Coci. Patients’ clinical and genetic data were acquired by Emanuele G. Coci, Maria Daniela D´Agostino, Laura Russell, Mouna Benamor, Alberto Fernández-Jaén, Ana Jimenez de Domingo, Lenka Noskova, Martin Magner, Sumit Parikh, Kimberly A Chapman, Himanshu Goel, Enrico Bertini, Ginevra Zanni, Ayman W. El-Hattab, Arif Khan, David Amor, A. Micheil Innes, Scott McLeod, Vanesa López-González, Maria J. Ballesta-Martinez, Nino Spataro, Carmen Manso, Maria Piccione, Emanuela Salzano, Myriam Srour, Sarah Alsubhi, Deborah Renaud, Martina Baethmann, Matias Wagner, Pablo Prieto Matos, Jill A. Rosenfeld, Daryl A. Scott, Syed A. Ahmed, Kristyn L. Rawson, Fróði Joensen, Marine Tessarech, Clement Proteau, Hong Li, David Michelson, Farzad Hashemi-Gorji, Audrey Putoux, Nicolas Chatron, Fanny Laffargue, Isabelle Creveaux, Rami Abou Jamra, Virginia Pironi, Hilde Van Esch, Tania Cruz Marino, Jessica Ricard, Trine Bjørg Hammer, Irene Valenzuela, Amaia Lasa-Aranzasti, Anna M. Cueto-Gonzalez , Purificación Marín Reina, Francisco Martinez , Milena Mariani , Claudia Ciaccio, Stefano D’arrigo , Annalaura Torella, Vincenzo Nigro, Manuela Morleo, Elsebet Ostergaard, Maria Palomares-Bralo andFernando Santos-Simarro. Statistical calculations were performed by Pilar Chacon Millan. DDX3X monomer modellingwas performed by Christoph G.W. Gertzen and Holger Gohlke. The first draft of the manuscript was written by Emanuele G. Coci, Christoph G.W. Gertzen and Holger Gohlke; all other authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Funding The Center for Structural Studies is funded by the DFG (Grant number 417919780). We are grateful for computational support and infrastructure provided by the “Zentrum für Informations- und Medientechnologie” (ZIM) at the Heinrich Heine University Düsseldorf and the computing time provided by the John von Neumann Institute for Computing (NIC) to HG on the supercomputer JUWELS at Jülich Supercomputing Centre (JSC) (user ID: VSK33, ETR1, DNAzyme). Multiple authors are members of the European Reference Network on Rare Congenital Malformations and Rare Intellectual Disability ERN-ITHACA [EU Framework Partnership Agreement ID: 3HP-HP-FPA ERN-01-2016/739516]. The work was supported by the project MULTIOMICS_CZ (Programme Johannes Amos Comenius, Ministry of Education, Youth and Sports of the Czech Republic,//ID Project CZ.02.01.01/00/23_020/0008540) – Co-funded by the European Union and by grant NU23-07-00281 from the Ministry of Health of the Czech Republic. Instrumental support was provided by The National Center for Medical Genomics (LM2023067). Ethics statement Parents or caregivers signed informed consent for publication of patients’ anonymized medical records. The study was approved by the Danish Patients Safety Authority (protocol number: 3-3013-2462/1 and 3-3013-3205/1) and the Capital Region of Denmark (protocol Rh-2018-20-6158). Competing interests The authors have nofinancial or non-financial interest to disclose. References Yedavalli VS, Neuveut C, Chi YH, Kleiman L, Jeang KT (2004) Requirement of DDX3 DEAD box RNA helicase for HIV-1 Rev-RRE export function. Cell 119:381–392 Von Mueffling A, Garcia-Forn M, De Rubeis S (2024) DDX3X syndrome: From clinical phenotypes to biological insights. J Neurochem 168(9):2147–2154 Snijders Blok L, Madsen E, Juusola J, Gilissen C, Baralle D, Reijnders MR, Venselaar H, Helsmoortel C, Cho MT, Hoischen A et al (2015) Mutations in DDX3X Are a Common Cause of Unexplained Intellectual Disability with Gender-Specific Effects on Wnt Signaling. Am J Hum Genet 97(2):343–352 Wang X, Posey JE, Rosenfeld JA, Bacino CA, Scaglia F, Immken L, Harris JM, Hickey SE, Mosher TM, Slavotinek A et al (2018) Phenotypic expansion in DDX3X – a common cause of intellectual disability in females. Ann Clin Transl Neurol 5(10):1277–1285 Lennox AL, Hoye ML, Jiang R, Johnson-Kerner BL, Suit LA, Venkataramanan S, Sheehan CJ, Alsina FC, Fregeau B, Aldinger KA et al (2020) Pathogenic DDX3X mutations impair RNA metabolism and neurogenesis during fetal cortical development. Neuron 106(3):404–420 Parra A, Pascual P, Cazalla M, Arias P, Gallego-Zazo N, San-Martin EA, Silván C, Santos-Simarro F, Spanish OverGrowth Registry Initiative (SOGRI), Nevado J et al (2024) Genetic and phenotypic findings in 34 Spanish patients with DDX3X neurodevelopmental disorder. Clin Genet 105(2):140-9 Kellaris G, Khan K, Baig SM, Tsai IC, Zamora FM, Ruggieri P, Natowicz MR, Katsanis N (2018) A hypomorphic inherited pathogenic variant in DDX3X causes male intellectual disability with additional neurodevelopmental and neurodegenerative features. Hum Genomics 12(1):11 Beal B, Hayes I, McGaughran J, Amor DJ, Miteff C, Jackson V, van Reyk O, Subramanian G, Hildebrand MS, Morgan AT et al (2019) Expansion of phenotype of DDX3X syndrome: six new cases. Clin Dysmorphol 28(4):169–174 Nicola P, Blackburn PR, Rasmussen KJ, Bertsch NL, Klee EW, Hasadsri L, Pichurin PN, Rankin J, Raymond FL, Study DDD et al (2019) De novo DDX3X missense variants in males appear viable and contribute to syndromic intellectual disability. Am J Med Genet A 179(4):570–578 Scala M, Torella A, Severino M, Morana G, Castello R, Accogli A, Verrico A, Vari MS, Cappuccio G, Pinelli M et al (2019) Three de novo DDX3X variants associated with distinctive brain developmental abnormalities and brain tumors in intellectually disabled females. Eur J Hum Genet 27(8):1254–1259 Dikow N, Granzow M, Graul-Neumann LM, Karch S, Hinderhofer K, Paramasivam N, Behl LJ, Kaufmann L, Fischer C, Evers C et al (2017) DDX3X mutations in two girls with a phenotype overlapping Toriello-Carey syndrome. Am J Med Genet A 173(5):1369–1373 Okano S, Miyamoto A, Makita Y, Taketazu G, Kimura K, Fukuda I, Tanaka H, Yanagi K, Kaname T (2020) Severe gastrointestinal symptoms caused by a novel DDX3X variant. Eur J Med Genet 63(12):104058 Chen Y, Liu KY, Yang ZL, Li XH, Xu R, Zhou H (2020) A de novo DDX3X variant is associated with syndromic intellectual disability: case report and literature review. Front Pediatr 8:303 Hu L, Xin X, Lin S, Luo M, Chen J, Qiu H, Ma L, Huang J (2020) A child with a novel DDX3X variant mimicking cerebral palsy: a case report. Ital J Pediatr 46(1):88 Moresco G, Costanza J, Santaniello C, Rondinone O, Grilli F, Prada E, Orcesi S, Coro I, Pichiecchio A, Marchisio P et al (2021) A novel de novo DDX3X missense variant in a female with brachycephaly and intellectual disability: a case report. Ital J Pediatr 47(1):81 Fieremans N, Van Esch H, Holvoet M, Van Goethem G, Devriendt K, Rosello M, Mayo S, Martinez F, Jhangiani S, Muzny DM et al (2016) Identification of Intellectual Disability Genes in Female Patients with a Skewed X-Inactivation Pattern. Hum Mutat 37(8):804–811 Sobreira N, Schiettecatte F, Valle D, Hamosh A (2015) GeneMatcher: A Matching Tool for Connecting Investigators with an Interest in the Same Gene. Hum Mutat 36(10):928–930 Song H, Ji X (2019) The mechanism of RNA duplex recognition and unwinding by DEAD-box helicase DDX3X. Nat Commun 10(1):1–8 Bas DC, Rogers DM, Jensen JH et al (2008) Very fast prediction and rationalization of pK(a) values for protein-ligand complexes. Proteins 73(3):765–783 Case DA, Metin Aktulnga H, Belfon K, Ben-Shalom I, Berryman JT, Brozell SR, Cerrutti DS, Cheatham TE, Cisneros GA, Cruzeiro VWD et al (2022) Amber22; University of California, San Francisco Wang J, Cieplak P, Kollman P (2000) How well does a restrained electrostatic potential (RESP) model perform in calculating conformational energies of organic and biological molecules? J Comput Chem 21(12):1049–1074 Zgarbova M, Otyepka M, Sponer J, Mladek A, Banas P, Cheatham TE, Jurecka P (2011) Refinement of the Cornell et al. nucleic acids force field based on reference quantum chemical calculations of glycosidic torsion profiles. J Chem Theory Comput 7(9):2886–2902 Banas P, Hollas D, Zgarbova M, Jurecka P, Orozco M, Cheatham TE, Sponer J, Otyepka M (2010) Performance of molecular mechanics force fields for RNA simulations: stability of UUCG and GNRA hairpins. J Chem Theory Comput 6(12):3836–3849 Perez A, Marchan I, Svozil D, Sponer J, Cheatham TE, Laughton C, Oroczo M (2007) Refinement of the AMBER force field for nucleic acids: improving the description of α/γ conformers. Biophys J 92(11):3817–3829 Humphrey W, Dalke A, Schulten K (1996) VMD: visual molecular dynamics. J Mol Graphics 14(1):33–38 Jumper J, Evans R, Prityel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool K, Bates R, Zidek A, Potapenko A et al (2021) Highly accurate protein structure prediction with AlphaFold. Nature 596(7873):583–589 Delgado J, Radusky LG, Cianferoni D, Serrano L (2019) FoldX 5.0: working with RNA, small molecules and a new graphical interface. Bioinformatics 35(20):4168–4169 Mann H, Whitney D (1947) Controlling the false discovery rate: A practical and powerful approach to multiple testing. Ann Math Stat 18(1):50–60 Wilcoxon F (1992) Individual comparisons by ranking methods. In: Kotz S, Johnson NL (eds) Breakthroughs in Statistics. Springer, New York, pp 196–202 Dai Y, Yang Z, Guo J, Li H, Gong J, Xie Y, Xiao B, Wang H, Long L (2022) Expansion of Clinical and Genetic Spectrum of DDX3X Neurodevelopmental Disorder in 23 Chinese Patients. Front Mol Neurosci 15:793001 Jayaraman K, Trachtmann N, Sprenger GA, Gohlke H (2022) Protein engineering for feedback resistance in 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase. Appl Microbiol Biotechnol 106(19–20):6505–6517 Babu MM (2016) The contribution of intrinsically disordered regions to protein function, cellular complexity, and human disease. Biochem Soc T 44(5):1185–1200 Jemth P, Karlsson E, Vogeli B, Guzovsky B, Andersson E, Hulquist G, Dogan J, Guntert P, Riek R, Chi CN (2018) Structure and dynamics conspire in the evolution of affinity between intrinsically disordered proteins. Sci Adv 4(10):eaau4130 Soulat D, Buerckstuemmer T, Westermayer S, Goncalves A, Bauch A, Stefanovic A, Hantschel O, Bennett KL, Decker T, Superti-Furga G (2008) The DEAD-box helicase DDX3X is a critical component of the TANK‐binding kinase 1‐dependent innate immune response. EMBO J 27(15):2135–2146 Tables Tables 1 and 2 are available in the Supplementary Files section. Supplementary Files Supplementarymaterialsandmethods.docx DDX3XTable1femalepatientsfinal.xlsx DDX3XTable2malepatientsfinal.xlsx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7842722","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":536966866,"identity":"eaf00323-7eb9-404f-b145-f18b0e92ab0d","order_by":0,"name":"Emanuele G. Coci","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqUlEQVRIiWNgGAWjYJCCAx8MbMAMCaK1HJxRkEaiFmaeD4dJ0MLf3vzwAI/Bebu1MxIYb3wgRovEmWMGByQMbidvO3OA2XIGMVoMJHIYDhgAtZgdb2CT5iFaS4LBuWSzwwxs0n+I1nLA4IAd2BZidID9crDBIDnB7MzBZsseYrQAQ+zx5z9/7OzNbiQfvPGDKGugILGBgbGBFA0MDPakKR8Fo2AUjIIRBQBf0jYk/a/QtAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0009-0009-3159-3754","institution":"Universitatsklinikum Magdeburg","correspondingAuthor":true,"prefix":"","firstName":"Emanuele","middleName":"G.","lastName":"Coci","suffix":""},{"id":536966867,"identity":"ee265b5c-3864-44c7-b371-ad50c4d3ad5b","order_by":1,"name":"Christoph G.W. Gertzen","email":"","orcid":"","institution":"Hochschule Dusseldorf","correspondingAuthor":false,"prefix":"","firstName":"Christoph","middleName":"G.W.","lastName":"Gertzen","suffix":""},{"id":536966868,"identity":"3d558f6e-be4e-4099-bc58-543d2501542d","order_by":2,"name":"Pilar Chacon Millan","email":"","orcid":"","institution":"TIGEM: Telethon Institute of Genetics and Medicine","correspondingAuthor":false,"prefix":"","firstName":"Pilar","middleName":"Chacon","lastName":"Millan","suffix":""},{"id":536966869,"identity":"7937aa30-2aa8-4684-b445-52c1e8de5e46","order_by":3,"name":"Maria Daniela D´Agostino","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Daniela","lastName":"D´Agostino","suffix":""},{"id":536966870,"identity":"cc492ff4-c18b-4518-9ef3-c16d6e1f0eb3","order_by":4,"name":"Laura Russell","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Laura","middleName":"","lastName":"Russell","suffix":""},{"id":536966871,"identity":"b8fbb832-e6dd-4027-9275-581527a032a3","order_by":5,"name":"Mouna Benamor","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Mouna","middleName":"","lastName":"Benamor","suffix":""},{"id":536966872,"identity":"266c304a-d2da-42ea-8ae7-5d859d5ea900","order_by":6,"name":"Alberto Fernández-Jaén","email":"","orcid":"","institution":"Universidad Europea: Universidad Europea de Madrid SLU","correspondingAuthor":false,"prefix":"","firstName":"Alberto","middleName":"","lastName":"Fernández-Jaén","suffix":""},{"id":536966873,"identity":"e1de5630-9ceb-49c2-82c3-2185cd56c002","order_by":7,"name":"Ana Jimenez de Domingo","email":"","orcid":"","institution":"Universidad Europea de Madrid Campus de Villaviciosa de Odón: Universidad Europea de Madrid SLU","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"Jimenez","lastName":"de Domingo","suffix":""},{"id":536966874,"identity":"3ffa7f3d-6f4a-4523-9660-f0226d2c7534","order_by":8,"name":"Lenka Noskova","email":"","orcid":"","institution":"Charles University First Faculty of Medicine: Univerzita Karlova 1 lekarska fakulta","correspondingAuthor":false,"prefix":"","firstName":"Lenka","middleName":"","lastName":"Noskova","suffix":""},{"id":536966875,"identity":"77dc49ec-5a6a-4bd3-8053-207623c42c3d","order_by":9,"name":"Martin Magner","email":"","orcid":"","institution":"Charles University First Faculty of Medicine: Univerzita Karlova 1 lekarska fakulta","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Magner","suffix":""},{"id":536966876,"identity":"780f89a5-69e2-4599-82c3-9b8eaf688a97","order_by":10,"name":"Sumit Parikh","email":"","orcid":"","institution":"Cleveland Clinic","correspondingAuthor":false,"prefix":"","firstName":"Sumit","middleName":"","lastName":"Parikh","suffix":""},{"id":536966877,"identity":"26054b27-df37-4ed2-b9c3-e1b2e93f1147","order_by":11,"name":"Kimberly A Chapman","email":"","orcid":"","institution":"Children's National Medical Center: Children's National Hospital","correspondingAuthor":false,"prefix":"","firstName":"Kimberly","middleName":"A","lastName":"Chapman","suffix":""},{"id":536966878,"identity":"3cb748f1-3875-442c-8047-52749f8d224b","order_by":12,"name":"Himanshu Goel","email":"","orcid":"","institution":"Hunter New England Health: Hunter New England Local Health District","correspondingAuthor":false,"prefix":"","firstName":"Himanshu","middleName":"","lastName":"Goel","suffix":""},{"id":536966879,"identity":"447b8e8e-2019-4981-a7ad-f93d3436c49b","order_by":13,"name":"Enrico Bertini","email":"","orcid":"","institution":"Bambino Gesu Pediatric Hospital: Ospedale Pediatrico Bambino Gesu","correspondingAuthor":false,"prefix":"","firstName":"Enrico","middleName":"","lastName":"Bertini","suffix":""},{"id":536966880,"identity":"9208aa53-48e9-474f-8555-c281baca3051","order_by":14,"name":"Ginevra Zanni","email":"","orcid":"","institution":"Bambino Gesu Pediatric Hospital: Ospedale Pediatrico Bambino Gesu","correspondingAuthor":false,"prefix":"","firstName":"Ginevra","middleName":"","lastName":"Zanni","suffix":""},{"id":536966881,"identity":"b9f7848a-8f63-4c1a-bbff-45a4ef60bffc","order_by":15,"name":"Ayman W. El-Hattab","email":"","orcid":"","institution":"University of Sharjah College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ayman","middleName":"W.","lastName":"El-Hattab","suffix":""},{"id":536966882,"identity":"90a3a44a-6a31-4bef-923a-67a0a93ba8ae","order_by":16,"name":"Arif Khan","email":"","orcid":"","institution":"Neuropedia children's neuroscience center","correspondingAuthor":false,"prefix":"","firstName":"Arif","middleName":"","lastName":"Khan","suffix":""},{"id":536966883,"identity":"8a4f4cdd-c5f0-4062-aac7-51bfa5b53618","order_by":17,"name":"David Amor","email":"","orcid":"","institution":"Royal Children's Hospital Research Institute: Murdoch Childrens Research Institute","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Amor","suffix":""},{"id":536966884,"identity":"9c1bc840-65e4-4263-9ce0-ce0849ac7e73","order_by":18,"name":"A. Micheil Innes","email":"","orcid":"","institution":"University of Calgary","correspondingAuthor":false,"prefix":"","firstName":"A.","middleName":"Micheil","lastName":"Innes","suffix":""},{"id":536966885,"identity":"bafd84f2-4005-4ad3-a117-1515290a2119","order_by":19,"name":"Scott McLeod","email":"","orcid":"","institution":"University of Calgary","correspondingAuthor":false,"prefix":"","firstName":"Scott","middleName":"","lastName":"McLeod","suffix":""},{"id":536966886,"identity":"65a4da92-8182-4676-979f-4d57c5d4d9f8","order_by":20,"name":"Vanesa López-González","email":"","orcid":"","institution":"Hospital Clinico Universitario Virgen de la Arrixaca","correspondingAuthor":false,"prefix":"","firstName":"Vanesa","middleName":"","lastName":"López-González","suffix":""},{"id":536966887,"identity":"531eceec-ec11-47d3-948d-74cb92fa2f79","order_by":21,"name":"Maria J. Ballesta-Martinez","email":"","orcid":"","institution":"Virgen de la Arrixaca University Hospital: Hospital Clinico Universitario Virgen de la Arrixaca","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"J.","lastName":"Ballesta-Martinez","suffix":""},{"id":536966888,"identity":"e639688f-1ac4-4b7e-aca3-107229402002","order_by":22,"name":"Nino Spataro","email":"","orcid":"","institution":"Universidad Autónoma de Barcelona: Universitat Autonoma de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Nino","middleName":"","lastName":"Spataro","suffix":""},{"id":536966889,"identity":"07676a1e-a58e-400d-b813-b8fd2c1c9437","order_by":23,"name":"Carmen Manso","email":"","orcid":"","institution":"Universidad Autónoma de Barcelona: Universitat Autonoma de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Carmen","middleName":"","lastName":"Manso","suffix":""},{"id":536966890,"identity":"130bbc6b-31f6-4199-ae83-94179e39415a","order_by":24,"name":"Maria Piccione","email":"","orcid":"","institution":"University of Palermo: Universita degli Studi di Palermo","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"","lastName":"Piccione","suffix":""},{"id":536966891,"identity":"bfc1e91e-93bd-4b80-9132-4e7addf375a1","order_by":25,"name":"Emanuela Salzano","email":"","orcid":"","institution":"Villa Sofia Cervello United Hospitals: Azienda Ospedaliera Ospedali Riuniti Villa Sofia Cervello","correspondingAuthor":false,"prefix":"","firstName":"Emanuela","middleName":"","lastName":"Salzano","suffix":""},{"id":536966892,"identity":"1db37f28-1e26-4a61-a5b6-fee13ed7145d","order_by":26,"name":"Myriam Srour","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Myriam","middleName":"","lastName":"Srour","suffix":""},{"id":536966893,"identity":"365d666a-1411-4dae-9647-50a18d3acf8b","order_by":27,"name":"Sarah Alsubhi","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Alsubhi","suffix":""},{"id":536966894,"identity":"6cd7685c-852b-45b1-bb00-ca7466767985","order_by":28,"name":"Deborah Renaud","email":"","orcid":"","institution":"Mayo Clinic College of Medicine and Science","correspondingAuthor":false,"prefix":"","firstName":"Deborah","middleName":"","lastName":"Renaud","suffix":""},{"id":536966895,"identity":"629353e6-270a-4006-b3e5-0decd8b192cb","order_by":29,"name":"Martina Baethmann","email":"","orcid":"","institution":"Klinikum Dritter Orden","correspondingAuthor":false,"prefix":"","firstName":"Martina","middleName":"","lastName":"Baethmann","suffix":""},{"id":536966896,"identity":"11518b4d-d887-49e5-beda-1df57019333d","order_by":30,"name":"Mathias Wagner","email":"","orcid":"","institution":"Technical University of Munich: Technische Universitat Munchen","correspondingAuthor":false,"prefix":"","firstName":"Mathias","middleName":"","lastName":"Wagner","suffix":""},{"id":536966897,"identity":"44fbbc6a-6fb3-47c8-9635-0a5fcce24780","order_by":31,"name":"Pablo Prieto Matos","email":"","orcid":"","institution":"University of Salamanca: Universidad de Salamanca","correspondingAuthor":false,"prefix":"","firstName":"Pablo","middleName":"Prieto","lastName":"Matos","suffix":""},{"id":536966898,"identity":"cfb6b826-351d-4a96-95c3-edf33716a4b6","order_by":32,"name":"Jill A. Rosenfeld","email":"","orcid":"","institution":"Baylor College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jill","middleName":"A.","lastName":"Rosenfeld","suffix":""},{"id":536966899,"identity":"98664df3-e49d-4afe-95eb-8cc7976d2baa","order_by":33,"name":"Daryl A. Scott","email":"","orcid":"","institution":"Baylor College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Daryl","middleName":"A.","lastName":"Scott","suffix":""},{"id":536966900,"identity":"bb9f177c-f7f5-4870-abcc-54b433c56b64","order_by":34,"name":"Syed A. Ahmed","email":"","orcid":"","institution":"Kaiser Permanente","correspondingAuthor":false,"prefix":"","firstName":"Syed","middleName":"A.","lastName":"Ahmed","suffix":""},{"id":536966901,"identity":"021b586a-ce67-4631-bf58-5f11b4933270","order_by":35,"name":"Kristyn L. Rawson","email":"","orcid":"","institution":"Kaiser Permanente","correspondingAuthor":false,"prefix":"","firstName":"Kristyn","middleName":"L.","lastName":"Rawson","suffix":""},{"id":536966902,"identity":"ae9bab54-a315-4cd7-95a1-3e39c61ebb93","order_by":36,"name":"Fróði Joensen","email":"","orcid":"","institution":"National Hospital of the Faroe Islands: Landssjukrahusid","correspondingAuthor":false,"prefix":"","firstName":"Fróði","middleName":"","lastName":"Joensen","suffix":""},{"id":536966903,"identity":"01313011-5abc-4c6f-a956-fcc39086d638","order_by":37,"name":"Marine Tessarech","email":"","orcid":"","institution":"Angers University: Universite d'Angers","correspondingAuthor":false,"prefix":"","firstName":"Marine","middleName":"","lastName":"Tessarech","suffix":""},{"id":536966904,"identity":"7a0a603e-8a0e-4bad-8d92-7d7985f12001","order_by":38,"name":"Clement Proteau","email":"","orcid":"","institution":"University of Angers: Universite d'Angers","correspondingAuthor":false,"prefix":"","firstName":"Clement","middleName":"","lastName":"Proteau","suffix":""},{"id":536966905,"identity":"5a2396fb-31b4-4205-b901-c4d5e83ac1e9","order_by":39,"name":"Hong Li","email":"","orcid":"","institution":"Emory University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hong","middleName":"","lastName":"Li","suffix":""},{"id":536966906,"identity":"1d068744-9023-410d-8e7f-dcdae967016c","order_by":40,"name":"David Michelson","email":"","orcid":"","institution":"Loma Linda University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Michelson","suffix":""},{"id":536966907,"identity":"598ded73-e6f8-4e7f-892c-0532a59122de","order_by":41,"name":"Farzad Hashemi-Gorji","email":"","orcid":"","institution":"Shahid Beheshti University of Medical Sciences School of Nursing and Midwifery","correspondingAuthor":false,"prefix":"","firstName":"Farzad","middleName":"","lastName":"Hashemi-Gorji","suffix":""},{"id":536966908,"identity":"51488eee-189c-457d-9004-d393af955f2e","order_by":42,"name":"Audrey Putoux","email":"","orcid":"","institution":"University Claude Bernard Lyon 1: Universite Claude Bernard Lyon 1","correspondingAuthor":false,"prefix":"","firstName":"Audrey","middleName":"","lastName":"Putoux","suffix":""},{"id":536966909,"identity":"3f4eec1e-3e90-4bd7-8314-11db8d0c96ac","order_by":43,"name":"Nicolas Chatron","email":"","orcid":"","institution":"University Claude Bernard Lyon 1: Universite Claude Bernard Lyon 1","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Chatron","suffix":""},{"id":536966910,"identity":"47721097-ec07-4323-9a3c-44aa832e71dc","order_by":44,"name":"Fanny Laffargue","email":"","orcid":"","institution":"Clermont-Ferrand I University: Universite Clermont Auvergne","correspondingAuthor":false,"prefix":"","firstName":"Fanny","middleName":"","lastName":"Laffargue","suffix":""},{"id":536966911,"identity":"ca03c207-0905-44fc-a5f2-f58fb14a7683","order_by":45,"name":"Isabelle Creveaux","email":"","orcid":"","institution":"Clermont-Ferrand I University: Universite Clermont Auvergne","correspondingAuthor":false,"prefix":"","firstName":"Isabelle","middleName":"","lastName":"Creveaux","suffix":""},{"id":536966912,"identity":"7799867d-4e2f-4839-926a-fde4a879d206","order_by":46,"name":"Rami Abou Jamra","email":"","orcid":"","institution":"Leipzig University: Universitat Leipzig","correspondingAuthor":false,"prefix":"","firstName":"Rami","middleName":"Abou","lastName":"Jamra","suffix":""},{"id":536966913,"identity":"3340fca5-1426-47f1-822b-17e0aa1dc773","order_by":47,"name":"Virginia Pironi","email":"","orcid":"","institution":"AUSL della Romagna: Azienda Unita Sanitaria Locale della Romagna","correspondingAuthor":false,"prefix":"","firstName":"Virginia","middleName":"","lastName":"Pironi","suffix":""},{"id":536966914,"identity":"69cde3d8-1a37-4532-a59e-9ab5f3e9aeaf","order_by":48,"name":"Hilde Van Esch","email":"","orcid":"","institution":"KU Leuven: Katholieke Universiteit Leuven","correspondingAuthor":false,"prefix":"","firstName":"Hilde","middleName":"Van","lastName":"Esch","suffix":""},{"id":536966915,"identity":"6bdd6465-4977-47b3-945f-6d1e1e7d2fe2","order_by":49,"name":"Cruz Marino Tania","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Cruz","middleName":"Marino","lastName":"Tania","suffix":""},{"id":536966916,"identity":"fa6f0b50-106d-4854-88af-852933630930","order_by":50,"name":"Jessica Ricard","email":"","orcid":"","institution":"CIUSSS de l'Estrie CHUS: Centre integre universitaire de sante et de services sociaux de l'Estrie Centre hospitalier universitaire de Sherbrooke","correspondingAuthor":false,"prefix":"","firstName":"Jessica","middleName":"","lastName":"Ricard","suffix":""},{"id":536966917,"identity":"4ac74be7-6b25-48c2-995b-a1a9b629f35a","order_by":51,"name":"Trine Bjørg Hammer","email":"","orcid":"","institution":"Copenhagen University Hospital: Rigshospitalet","correspondingAuthor":false,"prefix":"","firstName":"Trine","middleName":"Bjørg","lastName":"Hammer","suffix":""},{"id":536966918,"identity":"5db25636-6058-439b-9765-7fd967031154","order_by":52,"name":"Irene Valenzuela","email":"","orcid":"","institution":"Vall d'Hebron Hospital: Hospital Universitari Vall d'Hebron","correspondingAuthor":false,"prefix":"","firstName":"Irene","middleName":"","lastName":"Valenzuela","suffix":""},{"id":536966919,"identity":"2d6ed0db-9530-4b0b-8ba8-192a2559dfb1","order_by":53,"name":"Amaia Lasa-Aranzasti","email":"","orcid":"","institution":"HEB","correspondingAuthor":false,"prefix":"","firstName":"Amaia","middleName":"","lastName":"Lasa-Aranzasti","suffix":""},{"id":536966920,"identity":"980125e0-a8d1-484c-b42b-fa872e12391e","order_by":54,"name":"Anna M. Cueto-Gonzalez","email":"","orcid":"","institution":"Vall d'Hebron Hospital: Hospital Universitari Vall d'Hebron","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"M.","lastName":"Cueto-Gonzalez","suffix":""},{"id":536966921,"identity":"3df0fb0b-c4e6-4997-8238-f57800f3b9e0","order_by":55,"name":"Purificación Marín Reina","email":"","orcid":"","institution":"Hospital La Fe: Hospital Universitari i Politecnic La Fe","correspondingAuthor":false,"prefix":"","firstName":"Purificación","middleName":"Marín","lastName":"Reina","suffix":""},{"id":536966922,"identity":"30cc6ec3-489c-4bda-9355-22ac3b009816","order_by":56,"name":"Francisco Martinez","email":"","orcid":"","institution":"Hospital La Fe: Hospital Universitari i Politecnic La Fe","correspondingAuthor":false,"prefix":"","firstName":"Francisco","middleName":"","lastName":"Martinez","suffix":""},{"id":536966923,"identity":"46c27073-cf49-4f83-99a1-d2d0ac9c624e","order_by":57,"name":"Milena Mariani","email":"","orcid":"","institution":"Ospedale Sant'Anna di Como: Azienda Socio Sanitaria Territoriale Lariana","correspondingAuthor":false,"prefix":"","firstName":"Milena","middleName":"","lastName":"Mariani","suffix":""},{"id":536966924,"identity":"b9a04ca8-f182-41ec-ab08-85db55e5f4b8","order_by":58,"name":"Claudia Ciaccio","email":"","orcid":"","institution":"Istituto Nazionale Neurologico Carlo Besta: Fondazione IRCCS Istituto Neurologico Carlo Besta","correspondingAuthor":false,"prefix":"","firstName":"Claudia","middleName":"","lastName":"Ciaccio","suffix":""},{"id":536966925,"identity":"64edb9da-54bd-410c-a46f-f852f8ec8faf","order_by":59,"name":"Stefano D’arrigo","email":"","orcid":"","institution":"Istituto Nazionale Neurologico Carlo Besta: Fondazione IRCCS Istituto Neurologico Carlo Besta","correspondingAuthor":false,"prefix":"","firstName":"Stefano","middleName":"","lastName":"D’arrigo","suffix":""},{"id":536966926,"identity":"ab371791-f644-439f-bc8e-cf284c822ab0","order_by":60,"name":"Annalaura Torella","email":"","orcid":"","institution":"TIGEM: Telethon Institute of Genetics and Medicine","correspondingAuthor":false,"prefix":"","firstName":"Annalaura","middleName":"","lastName":"Torella","suffix":""},{"id":536966927,"identity":"0c3e0fc9-4587-4cfd-b152-42599778b3c6","order_by":61,"name":"Vincenzo Nigro","email":"","orcid":"","institution":"TIGEM: Telethon Institute of Genetics and Medicine","correspondingAuthor":false,"prefix":"","firstName":"Vincenzo","middleName":"","lastName":"Nigro","suffix":""},{"id":536966928,"identity":"3756b34b-dc29-4de3-a7d0-79ad317a8af5","order_by":62,"name":"Manuela Morleo","email":"","orcid":"","institution":"TIGEM: Telethon Institute of Genetics and Medicine","correspondingAuthor":false,"prefix":"","firstName":"Manuela","middleName":"","lastName":"Morleo","suffix":""},{"id":536966929,"identity":"b4ac872d-c036-4bdc-8995-2f695dfaebdf","order_by":63,"name":"Elsebet Ostergaard","email":"","orcid":"","institution":"University of Copenhagen: Kobenhavns Universitet","correspondingAuthor":false,"prefix":"","firstName":"Elsebet","middleName":"","lastName":"Ostergaard","suffix":""},{"id":536966930,"identity":"d6c076e9-0f93-4f02-95d4-d2b280329fd8","order_by":64,"name":"Holger Gohlke","email":"","orcid":"","institution":"University of Dusseldorf: Heinrich-Heine-Universitat Dusseldorf","correspondingAuthor":false,"prefix":"","firstName":"Holger","middleName":"","lastName":"Gohlke","suffix":""},{"id":536966931,"identity":"4a50f266-b803-48c5-bf13-3953afaa1fe2","order_by":65,"name":"Maria Palomares-Bralo","email":"","orcid":"","institution":"Universidad Autónoma de Madrid: Universidad Autonoma de Madrid","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"","lastName":"Palomares-Bralo","suffix":""},{"id":536966932,"identity":"05eedcc3-0764-4058-b329-c3a41665f05d","order_by":66,"name":"Fernando Santos-Simarro","email":"","orcid":"","institution":"Son Espases University Hospital: Hospital Universitario Son Espases","correspondingAuthor":false,"prefix":"","firstName":"Fernando","middleName":"","lastName":"Santos-Simarro","suffix":""}],"badges":[],"createdAt":"2025-10-12 19:42:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7842722/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7842722/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95662278,"identity":"68e7b8ed-1e9b-4260-832b-ee78b486384d","added_by":"auto","created_at":"2025-11-11 16:37:19","extension":"tiff","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/6933284a42e0b3e50676967c.tiff"},{"id":95662104,"identity":"69ad0c87-2f59-4638-9b7c-9687b77852b5","added_by":"auto","created_at":"2025-11-11 16:37:11","extension":"tiff","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/4dcb1b708f49864d0a581f9e.tiff"},{"id":95661962,"identity":"a4a50bdc-e4bf-4fc1-9d45-f3196b1470c7","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"tiff","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/63a083b669eb71ba04c9023a.tiff"},{"id":95661956,"identity":"9bf05a03-c4b8-4f14-b131-c2752a369772","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"tiff","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/3e40599795cd16a4def09641.tiff"},{"id":95662099,"identity":"b4d22563-c4ea-4b15-b881-f0d09b27f509","added_by":"auto","created_at":"2025-11-11 16:37:11","extension":"tiff","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/5a75d777bce4963b378ce241.tiff"},{"id":95661957,"identity":"45079dc1-8bfd-4921-bff0-64786d9d81f1","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"tiff","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/7e163918c7ff4a05fceffbdb.tiff"},{"id":95662116,"identity":"7137a102-e730-4350-a9a1-e6305cec5c63","added_by":"auto","created_at":"2025-11-11 16:37:12","extension":"tiff","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/8b432cb55f43a1287c812f03.tiff"},{"id":95661958,"identity":"ef0a30dd-c9c6-48b0-980a-eb3bcaca4a47","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"docx","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":287959,"visible":true,"origin":"","legend":"","description":"","filename":"Figure9.docx","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/5d40292b1916dcf63298f362.docx"},{"id":95797572,"identity":"05771f15-a9eb-4a6a-bcac-ae606832ca09","added_by":"auto","created_at":"2025-11-13 08:06:49","extension":"docx","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":693818,"visible":true,"origin":"","legend":"","description":"","filename":"figure8.docx","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/5cae11755b4d6b254c233658.docx"},{"id":95661993,"identity":"ef99af94-be80-4e54-bddc-2ee3b91f5f8d","added_by":"auto","created_at":"2025-11-11 16:37:03","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52745,"visible":true,"origin":"","legend":"","description":"","filename":"jmmeJMMED2500839.xml","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/85797020c4ccc2adc7d388cd.xml"},{"id":95661931,"identity":"6bf187aa-9ab1-499e-90c7-20e0d3b2c669","added_by":"auto","created_at":"2025-11-11 16:37:01","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1474,"visible":true,"origin":"","legend":"","description":"","filename":"JMMED250083916947.go.xml","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/9f9542e0c8ef7aafb8d6a878.xml"},{"id":95662276,"identity":"dfe1319f-5830-4c63-937b-3529e2b654fd","added_by":"auto","created_at":"2025-11-11 16:37:19","extension":"xml","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":833,"visible":true,"origin":"","legend":"","description":"","filename":"JMMED2500839Import.xml","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/25524d30bb7c783477d3899b.xml"},{"id":95662222,"identity":"2bd6430d-197f-4ba1-8b2e-407b73e53cc0","added_by":"auto","created_at":"2025-11-11 16:37:17","extension":"xml","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":182691,"visible":true,"origin":"","legend":"","description":"","filename":"JMMED25008390enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/e4ce5f0bc4d6a4f1480b0495.xml"},{"id":95662285,"identity":"72edf011-88f9-4013-b156-4682d927b161","added_by":"auto","created_at":"2025-11-11 16:37:19","extension":"tiff","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/e194094bda57f0a9805b5e33.tiff"},{"id":95662111,"identity":"ef78fdf5-f2a3-49e5-b93f-a6facfa0df9f","added_by":"auto","created_at":"2025-11-11 16:37:11","extension":"tiff","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/58c9c849c53ccc6bb3435103.tiff"},{"id":95662150,"identity":"7f4ce3aa-df40-4383-aece-b1e5b1ffb131","added_by":"auto","created_at":"2025-11-11 16:37:13","extension":"tiff","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/1255eec66fd568f2fc55b034.tiff"},{"id":95662233,"identity":"74e27d6e-c75c-4071-86bf-119dcfff8a28","added_by":"auto","created_at":"2025-11-11 16:37:18","extension":"tiff","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/4d31285f77c76e6bc37d9ff0.tiff"},{"id":95662171,"identity":"16838b1d-286b-44a7-8c22-49daf7266c53","added_by":"auto","created_at":"2025-11-11 16:37:14","extension":"tiff","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/fcdf7fa4fb3cc46535d51b15.tiff"},{"id":95662284,"identity":"6e097d48-61cd-43bd-8273-fe37625403fc","added_by":"auto","created_at":"2025-11-11 16:37:19","extension":"tiff","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/d07c921185e815a5b5d42acb.tiff"},{"id":95662130,"identity":"fc68d331-1187-4ede-af12-0cdd976a458f","added_by":"auto","created_at":"2025-11-11 16:37:12","extension":"tiff","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1558686,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7.tiff","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/be9b960dbc105fa3d18bc9a9.tiff"},{"id":95661762,"identity":"ef597cbc-98c3-406a-a82b-2c80e9a360d6","added_by":"auto","created_at":"2025-11-11 16:36:51","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":681153,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/7887cc9e8ba2eb563df9360e.png"},{"id":95662081,"identity":"13863f13-52cc-4914-bda1-9646bc50e553","added_by":"auto","created_at":"2025-11-11 16:37:10","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":275270,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/0d76096430046ce345a0e2e0.png"},{"id":95661953,"identity":"dde3e3d9-f089-408c-a09d-e92b14d06fc6","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26699,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/ba7181070c9ff51916bcb892.png"},{"id":95661940,"identity":"c68ea168-9d51-4638-b997-f0b12ec28b35","added_by":"auto","created_at":"2025-11-11 16:37:01","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18196,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/9a2926c15f7cbdd5647fb5ef.png"},{"id":95662157,"identity":"9d8cfcc7-96e1-454e-83d0-65a3d908b232","added_by":"auto","created_at":"2025-11-11 16:37:14","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23953,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/951d0c452217737c04c4bd53.png"},{"id":95662071,"identity":"9064cbbc-f276-46d2-9a48-298d48e183c8","added_by":"auto","created_at":"2025-11-11 16:37:09","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11395,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/f8e31f6c96b9f67b860083f7.png"},{"id":95662292,"identity":"994b9072-e031-4922-8cd5-ababb9df205f","added_by":"auto","created_at":"2025-11-11 16:37:20","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11897,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/ec2566f7268abc262d565d68.png"},{"id":95661783,"identity":"9e612663-7b25-4458-a0ac-8c449d612ca5","added_by":"auto","created_at":"2025-11-11 16:36:53","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9565,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/d035e39e908eb686938b4629.png"},{"id":95661967,"identity":"98c48a3c-4ed4-43d2-8422-178d6ebf1ab3","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22197,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure7.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/89ddbbec540eda0a7be91f61.png"},{"id":95661766,"identity":"511afe2e-3692-4f0e-8509-583df30a32a7","added_by":"auto","created_at":"2025-11-11 16:36:52","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102480,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/7d4fdacf10a9f6f6d77ec914.png"},{"id":95662122,"identity":"d391580a-038a-4306-9dda-afddc6ab1b6d","added_by":"auto","created_at":"2025-11-11 16:37:12","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56269,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/e75198bc1526ef5bd41792d2.png"},{"id":95662112,"identity":"84d21e45-2e74-43f9-b06f-425e1724b61c","added_by":"auto","created_at":"2025-11-11 16:37:11","extension":"xml","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":181544,"visible":true,"origin":"","legend":"","description":"","filename":"JMMED25008390structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/d9f517828c51472e6729b6bd.xml"},{"id":95661995,"identity":"6e376b95-9763-4cf2-b910-ee5e987112d0","added_by":"auto","created_at":"2025-11-11 16:37:03","extension":"html","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":215945,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/9f87892959618c39f3b95718.html"},{"id":95662118,"identity":"afa059f3-aba9-49ad-b842-598654936d2e","added_by":"auto","created_at":"2025-11-11 16:37:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":385557,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eDDX3X\u003c/em\u003e variants and DDX3X variations.\u003cstrong\u003e A \u003c/strong\u003e\u003cem\u003eDDX3X\u003c/em\u003e variants from 52 included female patients (black) and 7 included male patients (red) are depicted along \u003cem\u003eDDX3X\u003c/em\u003ecDNA. Splice site variants are underlined. Seven variants are reported twice (indicated in brackets). \u003cstrong\u003eB \u003c/strong\u003eOut of 59 patients, 35 patients (28 female and 7 male) display 31 amino acid variations, which are caused by 26 missense variants and 5 in-frame deletion variants. Amino acid variations are depicted black for 28 female and red for 7 male patients. Four amino acid variations are reported twice (indicated in brackets). ATP-binding helicase domain is depicted orange and C-terminal helicase domain green.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/920683a3b0928bcd5d4d0000.png"},{"id":95662064,"identity":"153dd994-885c-46d9-aa0f-944cec8ca430","added_by":"auto","created_at":"2025-11-11 16:37:08","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":292713,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eDDX3X\u003c/em\u003e variants \u003cstrong\u003eA\u003c/strong\u003eDistribution of variant types in female (blue) and male (yellow) patients. \u003cstrong\u003eB\u003c/strong\u003eDistribution of amino acid variation types in female patients. \u003cstrong\u003eC\u003c/strong\u003eLocalization of in-frame deletion and missense variants in female patients. \u003cstrong\u003eD\u003c/strong\u003eLocalization of missense variants in female and male patients. \u003cstrong\u003eE\u003c/strong\u003eVariants mapping in 3‘terminal gene tract in female and male patients.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/62f2e0aa12de6e95437a0d44.png"},{"id":95662044,"identity":"77875ea4-a607-4f9d-b9b9-cd87cfcbd4d8","added_by":"auto","created_at":"2025-11-11 16:37:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":540271,"visible":true,"origin":"","legend":"\u003cp\u003eSomatometric features \u003cstrong\u003eA\u003c/strong\u003e Variant types in female patients with weight, length and head circumference (hc) below the 10\u003csup\u003eth\u003c/sup\u003e centile, at birth as well as at the last follow-up. \u003cstrong\u003eB\u003c/strong\u003e Missense variants in female patients with weight, length and head circumference below the 10\u003csup\u003eth\u003c/sup\u003e centile, at birth as well as at last follow-up. \u003cstrong\u003eC\u003c/strong\u003e Localization of missense and in-frame deletion variants in female patients with weight, length and head circumference below the 10\u003csup\u003eth\u003c/sup\u003e centile, at birth as well as at last follow-up.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/2d98837582096df00b056469.png"},{"id":95662230,"identity":"68f9c9ee-a965-4790-b768-cfb724d40f6a","added_by":"auto","created_at":"2025-11-11 16:37:18","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":37214,"visible":true,"origin":"","legend":"\u003cp\u003eCranial, facial, skeletal and somatic dysmorphisms\u003cstrong\u003e A\u003c/strong\u003e Dysmorphologic subgroups 1 (score from 0 to 3), 2 (score from 4 to 6) and 3 (score from 7 to 10) in female and male patients. \u003cstrong\u003eB\u003c/strong\u003eLocalization of missense variants in female patients, divided into subgroups 1, 2 and 3 according to dysmorphology score.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/cfdb7e3968351745729da7dd.png"},{"id":95662282,"identity":"fcf205f1-ab51-486a-83c0-4f96f503da22","added_by":"auto","created_at":"2025-11-11 16:37:19","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":193133,"visible":true,"origin":"","legend":"\u003cp\u003eNeurological dysfunction\u003cstrong\u003e A\u003c/strong\u003e Distribution of variant types within neurological dysfunction subgroups 0 (normal motor development), 1 (isolated hypotonia), 2 (hypotonia plus one further neurological finding), 3 (hypotonia plus two or more neurological findings) of female cohort. \u003cstrong\u003eB\u003c/strong\u003e Localization of missense and in-frame deletion variants in female patients with and without neurological dysfunction.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/a1cec6b8055058d397f34f27.png"},{"id":95662385,"identity":"6bd7ea14-5cd7-4781-9108-8a80b31e2f56","added_by":"auto","created_at":"2025-11-11 16:37:27","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":154835,"visible":true,"origin":"","legend":"\u003cp\u003eEpilepsy \u003cstrong\u003eA\u003c/strong\u003e Distribution of variant types in female patients with and without epilepsy. \u003cstrong\u003eB\u003c/strong\u003e Localization of missense and in-frame deletion variants in female patients with epilepsy.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/19522ba45a276dc1d38cf91b.png"},{"id":95662003,"identity":"55fa3246-5110-424f-abdc-1a2b99fbf9fd","added_by":"auto","created_at":"2025-11-11 16:37:03","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":401687,"visible":true,"origin":"","legend":"\u003cp\u003eBrain MRI \u003cstrong\u003eA\u003c/strong\u003eDistribution of normal and pathologic brain MRI in variant types. \u003cstrong\u003eB\u003c/strong\u003eDistribution of variant types in normal and pathologic brain MRI subgroups of female patients. \u003cstrong\u003eC\u003c/strong\u003e Localization of missense and in-frame deletion variants in female patients with normal and pathologic brain MRI subgroups. \u003cstrong\u003eD\u003c/strong\u003eDistribution of variant types in female patients with pathologic brain MRI subgroup 1 (isolated ventricular enlargement or isolated corpus callosum anomalies) and subgroup 2 (multiple anomalies in two or more different brain regions). \u003cstrong\u003eE\u003c/strong\u003e Distribution of pathologic brain MRI subgroups 1 and 2 in female patients with missense and in-frame deletion variants.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/cd8a79bbaeec808f748f8ab4.png"},{"id":95662004,"identity":"cf4d341f-351f-4445-83ad-a3660efa060b","added_by":"auto","created_at":"2025-11-11 16:37:03","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":624502,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted effect and location of the variations in the DDX3X AlphaFold2 model \u003cstrong\u003eA\u003c/strong\u003ePredicted folding free energy (light green) as a measure of the impact on protein stability and the number of RNA still bound after MD simulations (dark green) as a measure of protein function for variations found in female (orange label) and male patients (navy label). \u003cstrong\u003eB\u003c/strong\u003eThe variations found in female patients (orange) are frequently found in the hydrophobic core of DDX3X and the variations found in male patients (navy) are mainly found on the protein surface. The location in the hydrophobic core facilitates misfolding as predicted via FoldX when small-to-large variations occur as identified in female patients.To improve clarity, only substitutions emphasized in the text are labeled. \u003cstrong\u003eC \u003c/strong\u003eRNA binding capability of mutant DDX3X monomers, divided into subgroups, in MD simulations. The number of RNA still bound (bronze) and unbound (brown) after MD simulations with significant differences indicated via solid lines and an asterisk (\u003cem\u003ep\u0026lt;0.05\u003c/em\u003e; Fisher’s exact test) or insignificant differences indicated via a dashed line and the actual \u003cem\u003ep\u003c/em\u003e-value. The values depicted over each bar were used for the statistical calculations.\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/fd7d9005973fee4bc73ed661.png"},{"id":95661981,"identity":"add8d967-81d5-430d-b9f0-2e287eabd53a","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":400727,"visible":true,"origin":"","legend":"\u003cp\u003eResults of perceptron models for categorizing clinical outcomes as “more severe” or “less severe” \u003cstrong\u003eA\u003c/strong\u003e F1 scores (y-axis) for perceptron models for different clinical outcomes in female as well as male patients and cutoffs (x-axis) in a stepsize of 0.1 within the severity score range (see text for a definition). The cutoff value assigns different weights to the categories “more severe” (above or equal to the cutoff) and “less severe” (below the cutoff) (see text).Generally, better models can be found for cutoffs \u0026gt; 0.6 except for male brain anomalies and male dysmorphisms. \u003cstrong\u003eB\u003c/strong\u003eROC-AUC scores for perceptron models as in panel A. \u003cstrong\u003eC \u003c/strong\u003eFor neurological dysfunction in female patients, the decision boundary computed via the perceptron (green dashed line) is shown, dividing the outcomes into “more severe” (blue dots) and “less severe” (red dots) cases (cutoff used: 0.7). “Less severe” cases are characterized by less severely impacted protein folding and protein function and vice versa. \u003cstrong\u003eD \u003c/strong\u003eFor epilepsy in female patients, the decision boundary computed via the perceptron (green dashed line) is shown, dividing the outcomes into “more severe” (blue dots) and “less severe” (red dots) cases (cutoff used: 0.7). “Less severe” cases are mainly characterized by less severely impacted protein function, whereas protein folding has a minor impact.\u003c/p\u003e","description":"","filename":"Figure9.png","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/7cc13ff24d3cd0e6edfb2958.png"},{"id":96919812,"identity":"01440e1b-bacf-4246-8dc6-527057a3f6ae","added_by":"auto","created_at":"2025-11-27 14:14:31","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4915487,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/9a2592ff-8b73-426a-bf70-d606f7202776.pdf"},{"id":95662123,"identity":"472df2ac-41aa-454a-86c7-bf4766b12f68","added_by":"auto","created_at":"2025-11-11 16:37:12","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":22007,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterialsandmethods.docx","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/e47d0f49dfb3c23538cc56f3.docx"},{"id":95661959,"identity":"f9f9df30-c667-4e37-8b5f-b364fc893846","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"xlsx","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":46845,"visible":true,"origin":"","legend":"","description":"","filename":"DDX3XTable1femalepatientsfinal.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/6944e3f9f03cb6eeffca6370.xlsx"},{"id":95661989,"identity":"aa97c061-6489-4857-83fb-87c313563f36","added_by":"auto","created_at":"2025-11-11 16:37:02","extension":"xlsx","order_by":16,"title":"","display":"","copyAsset":false,"role":"supplement","size":16945,"visible":true,"origin":"","legend":"","description":"","filename":"DDX3XTable2malepatientsfinal.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7842722/v1/fc1d38fd745e8f24142c899c.xlsx"}],"financialInterests":"","formattedTitle":"\u003cp\u003e\u003cem\u003eDDX3X \u003c/em\u003esyndrome: a multicenter genotype-phenotype study\u003c/p\u003e","fulltext":[{"header":"Key Message","content":"\u003cul\u003e\n \u003cli\u003eOur \u003cem\u003eDDX3X\u0026nbsp;\u003c/em\u003epatient cohort with 59 individuals is the second largest published ever\u003c/li\u003e\n \u003cli\u003eWe found correlations between variants’ type and position and phenotypical features\u003c/li\u003e\n \u003cli\u003eWe simulated mutant DDX3X monomers via AlphaFold2\u003c/li\u003e\n \u003cli\u003eWe compared mutant DDX3X monomers’ folding and RNA binding to wild-type DDX3X monomer\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"Introduction","content":"\u003cp\u003eThe clinical management of genetically determined neurodevelopmental disorders is a major effort for medical institutions, even more when the clinical presentation of a disorder is highly variable among affected individuals.\u003c/p\u003e\u003cp\u003eDEAD-box helicase 3 (\u003cem\u003eDDX3X\u003c/em\u003e) is member of DExD/H-box RNA helicase superfamily, maps in Xp11.4 band and encodes for an RNA-binding protein involved in entire RNA homeostasis [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. The helicase activity is performed by two enzymatic domains localized in the central part of the monomer, and the overall function is performed by a DDX3X homodimer.\u003c/p\u003e\u003cp\u003e\u003cem\u003eDDX3X\u003c/em\u003e syndrome was initially described in 2015 [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Since then, \u003cem\u003eDDX3X\u003c/em\u003e variants have been reported in a few studies [\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] and in several case reports [\u003cspan additionalcitationids=\"CR8 CR9\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]; clinical heterogeneity is clearly recognizable in \u003cem\u003eDDX3X\u003c/em\u003e deficiency: cognitive impairment spans from learning disability to profound intellectual disability and organic dysfunction is heterogeneous among patients [\u003cspan additionalcitationids=\"CR12 CR13 CR14\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn female cases, \u003cem\u003eDDX3X\u003c/em\u003e escapes X chromosome inactivation, and this phenomenon complicates the delineation of genotype-phenotype correlations [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. In the case of heterozygous missense and in-frame deletion variants, some female patients\u0026rsquo; cells simultaneously produce wild-type, mixed and mutant DDX3X homodimers in a total physiologic quantity; whereas heterozygous frameshift, stop-codon and splice site variants in female patients\u0026rsquo; cells exclusively produce wild-type DDX3X homodimers in reduced quantity.\u003c/p\u003e\u003cp\u003eMale patients harbor exclusively missense variants, whereas male embryos with loss of function variants die \u003cem\u003ein utero\u003c/em\u003e; with hemizygous missense variants, male patients\u0026rsquo; cells exclusively produce mutant DDX3X homodimers in physiologic amounts.\u003c/p\u003e\u003cp\u003eAlthough genetic, clinical and diagnostic data have been broadly reported, no study has focused on quantitative correlation between \u003cem\u003eDDX3X\u003c/em\u003e variant type and localization and clinical, psychometric and diagnostic findings. Based on 52 female and 7 male patients, we have studied the second largest \u003cem\u003eDDX3X\u003c/em\u003e patient cohort to date and have searched for quantitative correlations between underlying \u003cem\u003eDDX3X\u003c/em\u003e variant type and localization and phenotypic findings. As a first for \u003cem\u003eDDX3X\u003c/em\u003e neurodevelopmental syndrome, we performed molecular simulations for 24 DDX3X monomers, carrying single amino acid variations from 18 female and 6 male included missense variants, and classified the variations as to their clinical severity according to their folding free energy and RNA binding ability.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003ePatients\u003c/h2\u003e\u003cp\u003eWe identified and accrued female and male individuals with neurodevelopmental phenotype who harbor a \u003cem\u003eDDX3X\u003c/em\u003e variant using GeneMatcher [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] and through a search of the Baylor Genetics clinical database. We excluded individuals who carried pathogenic or likely pathogenic variants in other human disease genes. Our cohort consisted of fifty-two female and seven male patients whose medical data had been previously collected and recorded at the managing clinical institution during regular clinical interactions.\u003c/p\u003e\u003cp\u003e\u003cb\u003eDDX3X\u003c/b\u003e \u003cb\u003esequencing\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eDDX3X\u003c/em\u003e variants were identified in the probands using massively parallel sequencing (next generation sequencing) based technologies (exome/genome sequencing with or without employing virtual gene panels) in either clinical diagnostic or research settings. Parental testing was performed in most cases, either as part of trio exome/genome sequencing or by testing for the specific variant by Sanger sequencing. Data analysis, variant filtering, and prioritization were performed using the in-house implemented pipelines of the local genetic centers. Pathogenicity of the identified \u003cem\u003eDDX3X\u003c/em\u003e variants was established according to American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) criteria (Richards et al., 2015). All the variants are described based on the NM_001356.5 (GRCh37/hg19) transcript of \u003cem\u003eDDX3X\u003c/em\u003e in according to Human Genome Variation Society recommendations (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://varnomen.hgvs.org/\u003c/span\u003e\u003cspan address=\"https://varnomen.hgvs.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). All variants were confirmed using Mutalyzer (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://mutalyzer.nl\u003c/span\u003e\u003cspan address=\"https://mutalyzer.nl\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eContinuous and categorical variables for genotype-phenotype correlations were analyzed by performing Pearson Chi-square test in R version 4.3.2.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eDDX3X monomer modelling\u003c/h3\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003eModeling and Molecular Dynamics simulations\u003c/h2\u003e\u003cp\u003eThe wild-type structure of DDX3X in an RNA-bound state was generated by deleting one copy of DDX3X from the X-ray crystal structure (PDB-ID: 6O5F) [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Subsequently, gaps within the structure were filled using Prime and Maestro (\u003cem\u003eSchr\u0026ouml;dinger, LLC, New York, USA\u003c/em\u003e, 2017). The variations Ser181Thr, Phe182Val, Ile191Asn, Gln207Glu, Leu220Ser, Gly227Arg, Leu235Gln, Thr275Met, Arg326Cys, Arg326His, Arg351Gln, Arg376Cys, Ala404Asp, Gly406Arg, Trp421Gly, Leu427Gln, Leu484Pro, Arg488Cys, and Thr498Arg were created in Maestro, and for all variations and the wild-type the protonation states were assigned using PROPKA at pH 7.4 [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Molecular dynamics (MD) simulations were performed with Amber22 [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. The ff14SB force field was used to parameterize the protein, Joung-Chetham parameters were used for the counter ions, and TIP3P for the water. Here, the ff99 force field [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] with the \u0026ldquo;OL3\u0026rdquo; χ distribution from ff14SB [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] and the parmbsc0 α/γ [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] modifications resulting in the Amber ff99OL3 parameters were used for the RNA. For a description of how we generated input structures and thermalized the simulation systems, please see \u0026ldquo;Supplementary Methods\u0026rdquo;.\u003c/p\u003e\u003cp\u003eTwelve independent production runs of MD simulations with a constant number of particles, constant volume, and constant temperature (NVT) with 1 \u0026micro;s length each were performed. For this, the starting temperatures of the MD simulations at the beginning of the thermalization were varied by a fraction of a Kelvin. Unbinding of RNA was evaluated using the Virtual Molecular Dynamics (VMD) program [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. To assess the potentially detrimental effects of the variations on folding, we generated an AlphaFold2 [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] model of DDX3X to gain information about structural regions with variations not resolved in the X-ray crystal structure. This model was subsequently analyzed with FoldX [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. First, van der Waals clashes in the wild-type DDX3X were removed via the \u0026ldquo;optimize\u0026rdquo; command. Then, all residues in the protein were varied to all other residues and itself to judge the effect of the variation on folding. A repetition of this procedure revealed no differences between the computed energies for all variations.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eClassification of severity based on folding free energy and MD simulations of RNA binding\u003c/h3\u003e\n\u003cp\u003eTo classify the variations into those with a more severe or less severe clinical outcome, their folding free energy and their capability to have RNA remaining bound in MD simulations were used as descriptors. A perceptron was trained using a Python3.9 script; see \u0026ldquo;Supplementary Methods\u0026rdquo; for \u0026ldquo;Details for training a perceptron\u0026rdquo; as to the software libraries used. The script linearly scales the data from 0 to 1, where 1 means a better clinical outcome for the patient or a better-predicted function of DDX3X. Where no MD data was available, a function similar to the wild-type was assumed, as other strategies common in data preparation for machine learning, e.g., using the mean of the data set, resulted in worse predictions. Such variations are found in the unstructured termini of DDX3X, thus, they are unlikely to affect RNA binding. This is corroborated by no predictive models being found when assuming otherwise.\u003c/p\u003e\u003cp\u003eThe dataset was split into female and male patients, and for each group, the computational predictions were correlated with the severity score for each of four clinical outcomes (neurological dysfunction, epilepsy, brain anomalies, and dysmorphisms) via a perceptron using the \u0026ldquo;minimize\" function of SciPy. Here, \u0026ldquo;Nelder-Mead\u0026rdquo; was used as a method with 10,000 maximum iterations to fit the decision boundary for the perceptron. The severity score is transformed into a binary classifier during this step according to a cutoff, which is varied from 0 to 0.9 in steps of 0.1. Due to sparse training data (no. of data points per clinical outcome\u0026thinsp;\u0026lt;\u0026thinsp;7 for male and \u0026lt;\u0026thinsp;19 for female patients), no test set was used. All values equal to or below the cutoff are classified as class 0 (more severe) and all values above the cutoff are classified as class 1 (less severe). A cutoff of 1 would result in all severities classified as class 0 such that no predictions were possible.\u003c/p\u003e\u003cp\u003eTo train the perceptron, a linear decision boundary was fitted by minimizing a function that changes the slope and y-axis intercept of the decision boundary toward an optimal F1 score. See \u0026ldquo;Supplementary Methods\u0026rdquo; for \u0026ldquo;Details for line fitting\u0026rdquo;.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eDDX3X\u003c/strong\u003e \u003cstrong\u003evariants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur cohort includes 52 female (88%) and 7 male (12%) patients, including 1 pair of female siblings and 1 pair of male siblings. Out of 52 female patients, 47 harbor de novo variants, and 5 could not be tested for inheritance due to absent parental DNA samples. Out of 7 male patients, 2 harbor de novo variants and 5 maternally transmitted variants. Each patient harbors one single variant in \u003cem\u003eDDX3X\u003c/em\u003e, heterozygous in female and hemizygous in male individuals. In 52 female patients, we report 46 different variants: 20 missense (43.5%), 5 in-frame deletion (10.9%), 7 stop-gain (15.2%), 7 splice site (15.2%), 6 frameshift deletion (13.0%) and 1 frameshift duplication (2.2%); in 7 male patients, we report 6 different missense variants (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). Seven variants (3 missense, 2 stop-gain, 1 in-frame deletion, 1 frameshift) are reported twice. Out of 59 patients, 35 patients (28 female and 7 male) display 31 different amino acid variations, which are caused by 26 missense variants and 5 in-frame deletion variants; four amino acid variations are reported twice (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB).\u003c/p\u003e\n\u003cp\u003eTwenty-two out of 52 female patients harbor a missense variant and 2 missense variants were identified twice, and all 7 male patients harbor missense variants (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA). Out of 20 different missense variants (43.5%) from female patients, 18 (90%) cause a variation among different amino acid classes and 2 (10%) a variation within the same amino acid class (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eB). Out of 6 missense variants revealed in male patients, 3 (50%) cause a variation among different amino acid classes and 3 (50%) a variation within the same amino acid class.\u003c/p\u003e\n\u003cp\u003ePooling together 20 missense and 5 in-frame deletion variants from female patients, 18 (72%) map in exons coding for helicase domains and their binding tract, which span 55.2% of monomer length, and 7 (28%) map in exons coding N-terminal regions, which span 44.8% of monomer length (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eC). Out of 6 missense variants from male patients, 2 (33%) affect amino acids in a helicase domain, and 4 (66%) affect amino acids in the N-terminal or C-terminal regions (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eD). Twenty-four female patients harbor 14 loss of function (nonsense, stop-gain, frameshift deletion, frameshift duplication) variants and 7 splice site variants (together 45.7% of 46 different female variants). None of the 21 nonsense and splice site variants and none of the 25 missense and in-frame deletion variants from our female cohort map in 3` terminal \u003cem\u003eDDX3X\u003c/em\u003e portion, which encodes the C-terminal region (13.1% of monomer length) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eE).\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eClinical findings and correlation to underlying\u003c/strong\u003e \u003cstrong\u003eDDX3X\u003c/strong\u003e \u003cstrong\u003evariants\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eWe have analyzed the entire spectrum of clinical and diagnostic findings from 52 female and 7 male patients according to six categories (growth features, intellectual delay, epilepsy, brain anomalies, dysmorphisms, neurological findings); separately for female and male patients, we have searched for quantitative correlations between these phenotypic findings and underlying \u003cem\u003eDDX3X\u003c/em\u003e variant types and, for missense variants, class and localization of the mutated amino acid.\u003c/p\u003e\n\u003cp\u003eThe entire clinical and diagnostic data set for female patients is reported in Supplementary Table\u0026nbsp;1 (Suppl. Table\u0026nbsp;1) and for male patients in Supplementary Table\u0026nbsp;2 (Suppl. Table\u0026nbsp;2).\u003c/p\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n\u003ch2\u003eSomatometric features\u003c/h2\u003e\n\u003cp\u003eBirth somatometric parameters (weight, length, and head circumference) are fully reported for 38 out of 52 female patients; incomplete natal parameters are reported for 10 female patients and no natal parameters are reported for 4 female patients. Looking for a quantitative correlation between low prenatal growth and underlying \u003cem\u003eDDX3X\u003c/em\u003e variants, we have focused on patients with natal parameters below or equal to 10th centile. Thirteen patients with natal weight below or equal to 10th centile are reported, harboring 2 in-frame deletion variants (15.4%), 7 missense variants (53.8%) and 4 nonsense/splice site variants (30.8%); twelve patients with natal length below or equal to 10th centile are reported, harboring 7 missense variants (58.3%) and 5 nonsense/splice site variants (41.7%); twelve patients with natal head circumference below or equal to 10th centile are reported, harboring 3 in-frame deletion variants (25%), 5 missense variants (41.7%) and 4 nonsense/splice site variants (33.3%).\u003c/p\u003e\n\u003cp\u003eMeasurements for all three somatometric parameters at the last follow-up are reported for 46 out of 52 female patients. In 5 female patients, only two parameters are reported and in 1 female patient no parameter is reported. We have searched for a quantitative correlation between somatometric parameters below or equal to the 10th centile and underlying \u003cem\u003eDDX3X\u003c/em\u003e variants. A weight below or equal to 10th centile is reported in 15 patients, harboring 1 an in-frame deletion variant (6.7%), 11 a missense variant (73.3%) and 3 a nonsense/splice site variant (20%); a significant correlation is found between missense and in-frame deletion variants and low weight \u003cem\u003e(p\u0026thinsp;\u0026lt;\u0026thinsp;0.01)\u003c/em\u003e and between missense variants and low weight (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e). A length below or equal to the 10th centile is reported in 15 patients, harboring 3 in-frame deletion variants (20%), 10 missense variants (66.7%) and 2 nonsense/splice site variants (13.3%); a significant correlation is found between missense and in-frame deletion variant types and low length \u003cem\u003e(p\u0026thinsp;\u0026lt;\u0026thinsp;0.001)\u003c/em\u003e. A head circumference below or equal to the 10th centile is reported in 16 patients, harboring 1 in-frame deletion variant (6.3%), 9 missense variants (56.3%) and 6 nonsense/splice site variants (37.5%) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA and \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003c/p\u003e\n\u003cp\u003eIn female patients harboring a missense or in-frame deletion variant and displaying length below or equal to 10th centile at birth and at last follow-up, the variant is localized in helicases coding tracts more frequently than in the N-terminal region coding tract (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e for birth length; \u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e for last registered length). In female patients harboring a missense or in-frame deletion variant and displaying weight below or equal to 10th centile at birth, the variant is localized in helicases coding tracts more frequently than in the N-terminal region coding tract (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e). In female patients harboring a missense or in-frame deletion variant and displaying a head circumference below or equal to the 10th centile at birth and at the last follow-up, the variant is localized in the N-terminal region coding tract as prevalently as in helicases coding tracts (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eC).\u003c/p\u003e\n\u003cp\u003eOut of 7 male patients harboring missense variants, natal somatometric parameters are fully reported in 2, partially reported in 2, and not reported in 3 patients. Natal weight is reported in 4 patients, and none has a value below or equal to the 10th centile. Natal length is reported in 3 patients and none has a value below or equal to the 10th centile. Natal head circumference is reported in 3 patients and 1 patient has a value below 10th centile (33.3%). In 6 out of 7 male patients, somatometric parameters are also registered at the last follow-up. Weight is registered in 4 patients and is below 10th centile in 2 of them (50%). Length is registered in 4 patients and was below 10th centile in 2 of them (50%). Head circumference is registered in 5 patients and is below 10th centile in 1 of them (20%). In male patients displaying somatometric parameters below or equal to the 10th centile, a significant prevalence of localization of three missense variants (one in C-terminal helicase and two in non-enzymatic regions) in a specific \u003cem\u003eDDX3X\u003c/em\u003e tract cannot be revealed due to a small sample size.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\n\u003ch2\u003eCranial, facial, skeletal, and somatic dysmorphisms\u003c/h2\u003e\n\u003cp\u003eWe have quantified the dysmorphologic grade, searching for structural anomalies in ten body parts (skull, eyes, ears, philtrum/lips, palate, nose, chin, skeletal parts including hands and feet, internal organs, skin) and scoring each part with 1 point, if it displays at least one anomaly. Using a 10-point dysmorphology score, we have defined subgroup 1 (score from 0 to 3), subgroup 2 (score from 4 to 6) and subgroup 3 (score from 7 to 10).\u003c/p\u003e\n\u003cp\u003eDysmorphic findings are reported in 51 out of 52 female patients. Using the 10-point dysmorphology score, we have divided 51 female patients into subgroup 1 (31.4%), subgroup 2 (51.1%), and subgroup 3 (17.6%), revealing the different distribution of patients in the 3 subgroups to be statistically significant (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA).\u003c/p\u003e\n\u003cp\u003eAmong 16 patients in subgroup 1, 5 patients (31.2%) harbor nonsense variants, 8 patients (50%) missense variants, and 3 patients (18.7%) in-frame deletion variants. Among 26 patients in subgroup 2, 15 patients (57.7%) harbor nonsense variants, 10 patients (38.5%) missense variants, and 1 patient (3.8%) in-frame deletion variant. Among 9 patients in subgroup 3, 4 patients (44%) harbor nonsense variants, 4 patients (44%) missense variants, and 1 patient (12%) in-frame deletion variant. A direct correlation between dysmorphology score and variant types is not revealed; nevertheless, a direct correlation between dysmorphology score and localization of amino acid variations in helicase domains is revealed among patients with missense variants (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB).\u003c/p\u003e\n\u003cp\u003eDysmorphic findings are reported in all 7 male patients. Using the 10-point dysmorphology score, we have divided them into subgroup 1 (71.4%) and subgroup 2 (28.6%), and a correlation between dysmorphology score and missense variant localization is not evident, likely due to small sample size.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n\u003cdiv id=\"Sec15\" class=\"Section3\"\u003e\n\u003ch2\u003eNeurological dysfunction\u003c/h2\u003e\n\u003cp\u003eNeurological dysfunction score addresses muscle tone, muscle reflex, gait, balance and cranial nerves and was reported in 48 out of 52 female patients. We have divided the 48 female patients into four subgroups according to neurological dysfunction grade: 6 patients with normal motor development for age (score 0), 10 patients with isolated hypotonia (score 1), 13 patients with hypotonia plus one further neurological finding (score 2), and 19 patients with hypotonia plus two or more neurological findings (score 3).\u003c/p\u003e\n\u003cp\u003eMissense and in-frame deletion variants are seen in 15 out of 19 patients with score 3 (78.9%) and in 14 out of 29 patients with scores 0, 1, and 2 (48.3%); splice site/stop-codon/frame-shift variants are seen in 4 out of 19 patients with score 3 (21.1%) and in 15 out of 29 patients with score 0, 1 and 2 (51.7%). In female patients with neurological dysfunction grade 3, missense and in-frame deletion variants were revealed more frequently than nonsense and splice site variants (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eA).\u003c/p\u003e\n\u003cp\u003eOut of 25 female patients presenting neurological dysfunction of any grade and harboring missense and in-frame deletion variants, 6 patients (24%) have variants in N-terminal region coding tract and 19 patients (76%) in helicases coding tracts (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eB); all 3 female patients (100%) without neurological dysfunction have 2 missense variants in N-terminal region coding tract.\u003c/p\u003e\n\u003cp\u003eNeurological findings were reported in all 7 male patients; 4 patients were reported without any neurological dysfunction, 1 with a neurological dysfunction score 1 and 2 with a neurological dysfunction score 3. In the 3 male patients with neurological dysfunction, two missense variants map in N-terminal and C-terminal coding regions and do not lead to amino acid class change; one missense variant (Arg488Cys) maps in a helicase-coding region and leads to an amino acid class change.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n\u003cdiv id=\"Sec17\" class=\"Section3\"\u003e\n\u003ch2\u003eEpilepsy and EEG findings\u003c/h2\u003e\n\u003cp\u003eOut of 52 female patients, 10 have displayed epileptic seizures (19.2%) in the form of absences (6 patients), generalized seizures (4 patients), focal and complex-focal seizures (3 patients), and Blitz-Nick-Salaam (BNS) seizures (1 patient). In 10 female patients with epileptic seizures, missense variants (6 patients, 60%) and in-frame deletion variants (2 patients, 20%) are together more prevalent than splice site variants (2 patients, 20%) (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e). In 42 female patients without epileptic seizures, missense variants are seen in 16 patients (38.1%), in-frame deletion variants in 4 patients (9.5%), frameshift variants in 8 patients (19%), stop-gain variants in 9 patients (21.4%) and splice site variants in 5 patients (11.9%) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eA).\u003c/p\u003e\n\u003cp\u003eIn 8 female patients displaying epilepsy and harboring missense or in-frame deletion variants, 2 patients (25%) have the variant in the N-terminal region coding tract and 6 patients (75%) in gene tracts coding helicase domains and binding tract (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eB).\u003c/p\u003e\n\u003cp\u003eOut of 7 male patients, 2 patients have displayed epileptic generalized seizures (28.6%), and both associated amino acid variations (Arg603Gln and Gly607Ala) lie in the C-terminal region; since variation Gly607Ala is seen in one twin with and one twin without seizures, a correlation between epileptic seizures and C-terminal localization of this amino acid variation cannot be done.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n\u003cdiv id=\"Sec19\" class=\"Section3\"\u003e\n\u003ch2\u003eBrain anomalies\u003c/h2\u003e\n\u003cp\u003eOut of 52 female patients, 51 underwent brain MRI scans and are subdivided into two groups according to normal or pathologic findings of brain MRI. The prevalence of variant types (nonsense/splice site variants versus missense/in-frame deletion variants) within these 51 female patients is analyzed; twenty-three patients have nonsense or splice site variants, 7 (30.4%) displaying pathological brain MRIs and 16 (69.6%) normal brain MRIs (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e), and twenty-eight patients have missense and in-frame deletion variants, 20 (71.4%) displaying pathological brain MRIs and 8 (28.6%) normal brain MRIs (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eA).\u003c/p\u003e\n\u003cp\u003eAmong 24 female patients with normal brain MRI scans, 8 patients (33.3%) have missense and in-frame deletion variants and 16 patients (66.6%) have nonsense and splice site variants (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e); out of 8 female patients with normal brain MRI scans and missense or in-frame deletion variants, 6 patients (75%) have the variant in N-terminal region coding tract and 2 patients (25%) in helicase domains coding tracts. Among 27 female patients with pathologic brain MRI findings, 20 patients (74.1%) have missense or in-frame deletion variants and 7 patients (25.9%) have nonsense and splice site variants (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e); out of 20 female patients with pathologic brain MRI findings and missense or in-frame deletion variants, 3 patients (15%) harbor the variant in N-terminal region coding tract and 17 patients (85%) in tracts coding helicase domains and binding tract (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eB and \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eC).\u003c/p\u003e\n\u003cp\u003eAmong 27 female patients with pathological MRI findings, we distinguish 9 patients (subgroup 1) with isolated ventricular enlargement or isolated corpus callosum anomalies and 18 patients (subgroup 2) with multiple anomalies in two or more different brain regions. Within subgroup 1, 4 patients (44.4%) harbor nonsense variants and 5 patients (55.6%) missense and in-frame deletion variants; within subgroup 2, 3 patients (16.7%) harbor nonsense variants and 15 patients (83.3%) missense and in-frame deletion variants (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eD). Among 20 female patients displaying pathological brain MRI findings and harboring missense and in-frame deletion variants, 15 patients (75%) are included within subgroup 2 and 5 patients (25%) within subgroup 1 (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eE).\u003c/p\u003e\n\u003cp\u003eAmong 7 male patients, 6 patients underwent brain MRI studies; no brain anomalies are reported in 3 patients and pathological MRI findings are reported in 3 patients. In the 3 patients without brain anomalies in MRI scans, both underlying missense variants map within the C-terminal region; in the 3 patients with pathological MRI findings, 2 missense variants map in the N-terminal region and 1 missense variant in the ATP-binding helicase domain.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\n\u003cdiv id=\"Sec21\" class=\"Section3\"\u003e\n\u003ch2\u003eIntellectual disability\u003c/h2\u003e\n\u003cp\u003eOut of 52 female patients, IQ is measured in 20 patients by using one of the different standard psychometric tests and intellectual disability (ID) grade is diagnosed on measured IQ; IQ test is not performed in the remaining 24 patients and ID grade is estimated on several clinical observations. In 8 further patients, an ID grade cannot be reported by the managing clinician.\u003c/p\u003e\n\u003cp\u003eWe have gathered all 44 female patients with measured (20) ID as well as clinically estimated (24) ID, grouped them into three ID subgroups (mild-to-moderate ID, moderate-to-severe ID, severe ID) and looked for quantitative distribution of variant subsets within the 3 subgroups. Missense variants of female patients are more prevalent in the severe ID subgroup (50%) than in the other two ID subgroups (15% in mild-to-moderate ID subgroup and 35% in moderate-to-severe ID subgroup), being this different prevalence not statistically significant.\u003c/p\u003e\n\u003cp\u003eThe entire distribution of the variant subsets in the psychometric subgroups from measured and clinically estimated female patients is reported in \u0026ldquo;Supplementary Results\u0026rdquo;.\u003c/p\u003e\n\u003cp\u003eOut of 7 male patients, the ID grade could be estimated in 5 patients; among these 5 patients, IQ could be measured in 4 patients. No association can be revealed between ID grade and variations\u0026acute; position or between ID grade and variations\u0026acute; class change.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\n\u003ch2\u003eMolecular modeling and simulations of DDX3X variations\u003c/h2\u003e\n\u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\n\u003ch2\u003eFemale variations are predicted to be more detrimental to DDX3X folding than male variations\u003c/h2\u003e\n\u003cp\u003eTo test our hypothesis that variations found in female patients are more detrimental to protein function than variations found in male patients, we used a structural model created via AlphaFold2 [\u003cspan class=\"CitationRef\"\u003e26\u003c/span\u003e] to predict the effect on the free energy of folding via FoldX [\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e]; positive values indicate that the variation is less stable than the wild-type.\u003c/p\u003e\n\u003cp\u003eThe variations found in males Gly117Val, Ser181Thr, Arg351Gln, Arg488Cys, Arg603Gln, and Gly607Ala and the variations found in females Pro40Leu, Pro41His, Phe182Val, Ile191Asn, Gln207Glu, Leu220Ser, Gly227Arg, Leu235Gln, Thr275Met, Arg326Cys, Arg326His, Arg376Cys, Ala404Asp, Gly406Arg, Trp421Gly, Lys427Gln, Leu484Pro, and Thr498Arg were analyzed. The highest detrimental effect on the DDX3X folding free energy was 2.34 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (R603Q) among variations from males and 6.44 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Gly227Arg; Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003eA) among variations from females with a mean folding free energy over all variations from males of 0.63 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e and over all variations from females of 2.55 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. Five out of the six variations found in male patients led to a folding free energy change lower than 1.4 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, whereas the same was predicted for only three out of 19 variations found in female patients. The cutoff of 1.4 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e was used as this signifies a ten-times lower likelihood of being correctly folded compared to the wild-type. Comparing the detrimental energies of the male and female cohorts with a Mann-Whitney U test [\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e] reveals a statistically significant difference (\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.002\u003c/em\u003e) between the two, which shows that the variations found in female patients are significantly more detrimental to protein folding than the variations found in males. The male variations are mainly (five of six) located on the protein surface, whereas many of the female variations appear in the hydrophobic protein core (eight of 18) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003eB). This is also true for the most detrimental female variation Gly227Arg, which increases the size of the sidechain by six heavy atoms; several variations from females are small-to-large changes in the sidechain. Yet, also variations with a negative (i.e., favorable compared to the wild-type) change in the folding free energy were identified. Still, these variations might impact the RNA binding capability or the helicase activity of DDX3X.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\n\u003cdiv id=\"Sec25\" class=\"Section3\"\u003e\n\u003ch2\u003eMD simulations suggest reduced RNA binding in DDX3X variations\u003c/h2\u003e\n\u003cp\u003eAs some variations might lead to reduced RNA binding even if properly folded, we performed MD simulations of the DDX3X wild-type and the Ser181Thr, Phe182Val, Ile191Asn, Gln207Glu, Leu220Ser, Gly227Arg, Leu235Gln, Thr275Met, Arg326Cys, Arg326His, Arg351Gln, Arg376Cys, Ala404Asp, Gly406Arg, Trp421Gly, Leu427Gln, Leu484Pro, Arg488Cys, and Thr498Arg variations in the presence of a bound RNA double strand. Not all variations analyzed in the previous section could be included in the MD analysis, as their locations were not resolved in the X-ray crystal structure with bound RNA. The X-ray crystal structure of DDX3X co-crystallized with RNA was used (PDB ID 605F\u003csup\u003e1\u003c/sup\u003e) to generate starting structures for the wild-type and the variations. In the MD simulations, the RNA remained bound to the DDX3X wild-type in seven out of twelve replicas. Variations in the DDX3X monomer from all patients (\u003cem\u003ep\u0026thinsp;=\u0026thinsp;0.04\u003c/em\u003e), from female patients (\u003cem\u003ep\u0026thinsp;=\u0026thinsp;0.04\u003c/em\u003e), with folding free energy\u0026thinsp;\u0026lt;\u0026thinsp;1.4 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (\u003cem\u003ep\u0026thinsp;=\u0026thinsp;0.04\u003c/em\u003e), with negative folding free energy (\u003cem\u003ep\u0026thinsp;=\u0026thinsp;0.06\u003c/em\u003e), and from male patients (\u003cem\u003ep\u0026thinsp;=\u0026thinsp;0.07\u003c/em\u003e) confer a reduced RNA binding property compared to DDX3X wild-type (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003eC). Thus, MD simulations show that RNA is less likely to stay bound in mutated DDX3X than in DDX3X wild-type; this reduction is statistically significant in the first three subgroups.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePredictions of folding free energy and RNA binding can be used to classify variation outcomes into more severe and less severe clinical cases\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe hypothesized that variations found in female patients have a more detrimental effect on protein folding and function than those found in male patients, and our computational predictions correlate well with this hypothesis.\u003c/p\u003e\n\u003cp\u003eWe thus probed next whether the predicted effects also correlate with the severity of the assessed clinical outcomes. To do so, we trained perceptron models with a linear decision boundary towards an optimal F1-score. Initially, we encoded the clinical outcomes reported above into a linearized severity score ranging from 0 to 1. For neurological dysfunction, a severity score of 0 (best outcome) corresponds to a clinical score of 0 (normal motor development) and a severity score of 1 (worst outcome) corresponds to a clinical score of 3 (hypotonia plus two or more neurological findings). Choosing a cutoff within the severity score range provides a binary classification where patients with scores above and including the cutoff are treated as \u0026ldquo;more severe\u0026rdquo; and those with scores below the cutoff are treated as \u0026ldquo;less severe\u0026rdquo;. This allows us to assign different weights to the categories \u0026ldquo;more severe\u0026rdquo; and \u0026ldquo;less severe\u0026rdquo;. For the above mentioned example of neurological dysfunction, a cutoff of 0.7 would only classify variants with a clinical score of 2 to 3 as \u0026ldquo;more severe\u0026rdquo;.\u003c/p\u003e\n\u003cp\u003eFor four clinical outcomes (neurological dysfunction, epilepsy, brain anomalies, dysmorphisms) in female as well as male patients, a perceptron model was trained using the predicted protein folding and function data as input to best categorize the patients. The goodness of the categorization is assessed by F1 and ROC-AUC scores; F1 score ranges from 0 (no classification) to 1 (perfect classification) and includes both precision and recall, ROC-AUC score ranges from 0 (inverse perfect classification) through 0.5 (random classification) to 1 (perfect classification). The perceptron models yielded F1 scores up to 1.0 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eA) and ROC-AUC scores up to 1.0 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eB). Generally, models with the highest F1 scores are based on severity score cutoffs between 0.7 and 1. Yet, no satisfying classifications for the outcomes \u0026ldquo;brain anomalies\u0026rdquo; and \u0026ldquo;dysmorphisms\u0026rdquo; in male patients could be identified for the cutoffs 0.5 to 1; for lower cutoffs, models with F1-scores\u0026thinsp;~\u0026thinsp;0.8 are found.\u003c/p\u003e\n\u003cp\u003eExemplary categorization results into \u0026ldquo;more severe\u0026rdquo; and \u0026ldquo;less severe\u0026rdquo; cases, based on the combination of protein function and folding properties, are shown together with the decision boundary determined by the perceptron (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eC and \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eD). The model for neurological dysfunction in female patients shows that the decision boundary (green dotted line) is close to the line of ideal anticorrelation with respect to protein function and folding; it means that less severe cases are characterized by good protein function and folding and vice versa (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eC). The model for epilepsy in female patients (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eD) shows a decision boundary with a lower slope than in the model for neurological dysfunction (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eC), indicating that the protein folding properties are less determining for classification.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003cem\u003eDDX3X\u003c/em\u003e syndrome is one of the most prevalent genetic neurodevelopmental disorders. Defining robust correlations between underlying \u003cem\u003eDDX3X\u003c/em\u003e variants and phenotypic findings would sharpen the clinical prognostic view after genetic diagnosis and might facilitate the entire clinical management.\u003c/p\u003e\u003cp\u003eusing a cohort of 52 female and 7 male patients, our study analyses the second-largest \u003cem\u003eDDX3X\u003c/em\u003e patient cohort published so far. By considering type and localization of identified \u003cem\u003eDDX3X\u003c/em\u003e variants and the clinical and diagnostic findings of affected patients, we have delineated novel features caused by \u003cem\u003eDDX3X\u003c/em\u003e variants and have revealed previously unidentified correlations between \u003cem\u003eDDX3X\u003c/em\u003e genotypes and specific pathophenotypes.\u003c/p\u003e\u003cdiv id=\"Sec28\" class=\"Section2\"\u003e\u003ch2\u003eVariant types and localization and class of variations differ among sexes\u003c/h2\u003e\u003cp\u003eThe subset of missense and in-frame deletion variants as well as the subset of stop-gain, frameshift and splice site variants are revealed in similar proportions within our female cohort, showing that a quantitatively normal population of wild-type, mixed and mutant DDX3X homodimers as well as a reduced population of wild-type DDX3X homodimers are both pathogenic. While no variants map in the final two exons (16 and 17) in our female patients, a few stop-gain variants (Tyr576*, Arg602*, Arg603*) and one mosaic missense variant (Arg602Gln) were reported in those two exons in previously reported female patients [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. It might imply that dysfunction of one single C-terminal region is acceptable in mixed DDX3X homodimer and that almost all female individuals harboring \u003cem\u003eDDX3X\u003c/em\u003e missense variants in exons 16 and 17 are asymptomatic and remain undiagnosed.\u003c/p\u003e\u003cp\u003eOur study includes the largest male \u003cem\u003eDDX3X\u003c/em\u003e cohort analyzed to date. Five novel amino acid variations (Gly117Val, Ser181Thr, Arg488Cys, Arg603Gln, Gly607Ala) enlarge the current view on male \u003cem\u003eDDX3X\u003c/em\u003e genetics [\u003cspan additionalcitationids=\"CR4 CR5\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Out of six male missense variants, two are \u003cem\u003ede novo\u003c/em\u003e and four are inherited from healthy mothers, implying that the effect of these missense variants of hemizygous male patients is too weak in heterozygous females to cause a manifest disease. In line with the paradigmatic healthiness of variant-transmitting heterozygous mothers from our and previous \u003cem\u003eDDX3X\u003c/em\u003e studies, missense variants of our male patients were not revealed in our female patients. Nevertheless, our \u003cem\u003ede novo\u003c/em\u003e male variant Arg488Cys was reported in a previous female patient [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] and this is, to our knowledge, the first overlap of one \u003cem\u003eDDX3X\u003c/em\u003e variant among a male and a female patient; based on this first observation, we firstly question the above-mentioned paradigm.\u003c/p\u003e\u003cp\u003eWhile no variation has been reported in C-terminal region in a male before, our two novel variations Arg603Gln and Gly607Ala map distal to C-terminal helicase domain and this shed new light on function of C-terminal region in mutant DDX3X homodimer.\u003c/p\u003e\u003cp\u003eRegarding the localization of variations, missense variants from our patients differ among sexes; female variations are more prevalent in helicase domains (14) than in N-terminal region (5) and, contrarily, male variations are more prevalent in N- and C-terminal regions (4) than in helicase domains (2). Merging 11 male variations (9 in helicase domains and 2 in N-terminal region) from former studies with our 5 novel male variations, we note a higher prevalence in helicase domains (11 variations) than in N- and C-terminal regions (6 variations) like for female patients, although the ratio of variations in enzymatic tracts to variations in not-enzymatic tracts is less marked in males than in females.\u003c/p\u003e\u003cp\u003eLooking at the class of variations in our female and male patients, they clearly differ among sexes; in female patients, variations into a different amino acid class are 9 times more frequent than those into the same amino acid class, probably because variations of a wild-type amino acid into a similar one does not impair the function of mixed DDX3X homodimer strongly enough to cause a clinical pathophenotype. This female discrepancy in amino acid class change is not observed in our male patients, who display amino acid variations into the same class as frequently as into another class, and it likely indicates that each mutated DDX3X homodimer from a hemizygous variant is pathogenic. Uniting 10 former male variations (8 within the same class and 2 into another class) together with our 5 novel male variations, we note a higher prevalence of change into another amino acid class (11) than within the same one (5) like for female patients, although the male ratio is lower than in female patients.\u003c/p\u003e\u003cp\u003e\u003cb\u003eNovel correlations in our female patients and an association trend in our male patients emerge between DDX3X genotype and specific pathophenotypes\u003c/b\u003e\u003c/p\u003e\u003cp\u003eIn our female cohort, missense and in-frame deletion variants are more prevalently associated with low weight and length than nonsense and splice site variants; this gap increases from natal age to last follow-up age and might indicate that mutant and mixed DDX3X homodimers have a more detrimental impact on corporal growth than decreased wild-type DDX3X homodimers. Somatometric centiles decrease also in males over time. This phenomenon, firstly described in our study, might imply that \u003cem\u003eDDX3X\u003c/em\u003e has a stronger role in cell proliferation during postnatal growth than during fetal period.\u003c/p\u003e\u003cp\u003eA dysmorphology score for \u003cem\u003eDDX3X\u003c/em\u003e syndrome was first established in our study. We have observed that a direct correlation between dysmorphology score and localization of missense variants in helicase coding regions characterizes female patients; in male patients, no correlation between dysmorphology score and localization of missense variants has been identified.\u003c/p\u003e\u003cp\u003eWe first implemented a motor-focused neurological score in \u003cem\u003eDDX3X\u003c/em\u003e clinical research and correlated it to underlying \u003cem\u003eDDX3X\u003c/em\u003e zygosity. Two interesting neurological observations have risen up in our female cohort: first, the more severe the neurological dysfunction, the more frequent a missense or in-frame deletion variant; second, females with a missense variant in helicase-coding regions always display a neurological dysfunction, and this is relevant for clinical management and genetic consulting. Since neurological findings were not or were partially described in former male patients, previous data cannot be used to depict any correlation between male neurological dysfunction grade and variations\u0026rsquo; features, and our male patients\u0026rsquo; data are unconclusive in this regard.\u003c/p\u003e\u003cp\u003eEpilepsy is much more prevalent in \u003cem\u003eDDX3X\u003c/em\u003e patients (about 20%) than in the general population, considering together female and male patients from our study. Since our female epileptic patients harbor missense and in-frame deletion variants more prevalently than nonsense variants and since missense variants are more frequent in helicase coding regions than in N-terminal coding regions in this patient subgroup, it might imply that mutant and mixed DDX3X homodimers cause more severe electrophysiological neuronal dysfunction than quantitatively reduced wild-type DDX3X homodimers. As for neurological findings, our data on male epilepsy do not lead to any correlative conclusion.\u003c/p\u003e\u003cp\u003eTwo correlations between variant type and brain MRI findings and, respectively, localization of missense and in-frame deletion variants and brain MRI findings have emerged in our female cohort. Grouping female patients with pathologic MRI findings in two severity subgroups (1 and 2), we have observed that severe findings (subgroup 2) are more prevalently associated to missense and in-frame deletion variants than to nonsense variants, as observed in a former study focusing on polymicrogyria [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]; it implies that mutated and mixed DDX3X homodimers cause more severe structural brain anomalies than quantitatively reduced wild-type DDX3X homodimers. A clear correlation between localization of missense and in-frame deletion variants in helicase domains and pathologic brain MRI has risen up as well. Although a trend between variant localization in N-terminal region and pathologic brain findings can be observed in our male patients, the low number of brain MRIs does not allow any statistical conclusion, and future male \u003cem\u003eDDX3X\u003c/em\u003e patients will show whether proximal position of missense variants is or isn\u0026rsquo;t associated to structural brain anomalies.\u003c/p\u003e\u003cp\u003eIntellectual disability is a well-defined phenotypic feature of \u003cem\u003eDDX3X\u003c/em\u003e syndrome. In our female patients, missense variants were observed more prevalently in the severe ID subgroup (50%) than in the moderate-to-severe (35%) and mild-to-moderate (15%) ID subgroups. Although this trend is biased by an unclear delimitation of ID subgroups\u0026rsquo; borders and has to be analyzed more precisely in future \u003cem\u003eDDX3X\u003c/em\u003e studies, it might imply that mutant and mixed DDX3X homodimers cause a more severe neuronal dysfunction than reduced wild-type DDX3X homodimers and therefore give rise to severe ID grade. In our male patients, no correlation between genotype features and phenotype findings has been noted.\u003c/p\u003e\u003cp\u003eTaken together the entire female data, missense and in-frame variant types as well as localization of amino acid variations in helicase domains are directly correlated to a severe grade in five of six analyzed pathophenotypic findings. Considering the entire male data, one association trend rises up between missense variant localization and brain anomalies.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec29\" class=\"Section2\"\u003e\u003ch2\u003eMolecular simulations identify sex-specific effects in DDX3X variations\u003c/h2\u003e\u003cp\u003eThe three-times higher prevalence of DDX3X variations in female patients compared to male patients suggests that more variant-harbouring male embryos die \u003cem\u003ein utero\u003c/em\u003e than females. Hence, we hypothesized that variations in females cause DDX3X to be non-functional whereas variations in males only reduce the functionality of DDX3X. We used FoldX to predict how severe the impact of a variation on folding is compared to wild-type [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], as a misfolded protein will not be functional. In the FoldX analysis, variation-bearing monomers in females had on average a 1.92 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e higher detrimental energy, i.e., they are on average 25 times less likely to correctly fold than variation-bearing monomers in males. The average variation in males is only three times less likely to be correctly folded compared to wild-type. A variation from a female leading to strong misfolding likely leads to death when occurring in a male embryo and, therefore, we have not found any overlap between our female and male variations. Nevertheless, the first overlap of a variation (Arg488Cys, 0.79 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) between our male patient and a former female patient might raise the hypothesis that a variation can cause a clinical pathophenotype in both sexes if the associated folding free energy is not too high, allowing the viability of a male embryo, and not too low, causing clinically relevant findings in a female individual. According to this hypothesis, male variation Arg603Gln (2.34 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) should be revealed also in future female patients and its report in our current male patient might be clarified through its localization in a not-enzymatic region. In female patients with a wild-type copy, the phenotype should be less drastic as at least a quarter of all DDX3X dimers should be folded correctly.\u003c/p\u003e\u003cp\u003eMost of the variations from males and some variations from females resulted in monomer folding free energies\u0026thinsp;\u0026lt;\u0026thinsp;1.4 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e implying that their folding behavior is not much different from the DDX3X wild-type. Thus, we hypothesized that these variations should reduce the functionality of DDX3X, e.g., by decreasing the RNA binding capability. To probe this, we performed MD simulations of one DDX3X monomer (see below) in its folded state in the presence of RNA. We included all variations that could be mapped to an X-ray crystal structure of DDX3X in a complex with RNA (PDB ID: 6O5F) [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] and found that DDX3X variations are significantly less likely to stay bound to RNA compared to the wild-type. This holds true for variations with a FoldX energy\u0026thinsp;\u0026lt;\u0026thinsp;1.4 kcal mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, which are likely to fold correctly; hence, these variations likely hamper the functionality of DDX3X by lowering its affinity towards RNA rather than leading to a misfolded protein.\u003c/p\u003e\u003cp\u003eWe used a monomeric form of DDX3X bound to RNA in the MD simulations as monomers bind to dsRNA such that each protomer recognizes one RNA strand, i.e., the RNA binding is independent of a dimer state [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The dimers then cooperate in the unwinding of the dsRNA [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Hence, this setup allowed us to study the RNA binding capabilities of DDX3X variations in a time-efficient manner. We did not study deletion variations in our computational approaches as the exact effect of the deletion on the structure is difficult to predict. Finally, several variations, e.g., Pro40Leu, Pro41His, Arg603Gln, and Gly604Ala, are located in the intrinsically disordered regions (IDRs) of the C- and N-terminus of DDX3X; as IDRs often convey interactions between proteins [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] and DDX3X has several non-conventional functions facilitated by interaction with other proteins [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], the variations may influence these non-canonical functions.\u003c/p\u003e\u003cp\u003eFinally, we identified two models to classify patients into more severe and less severe outcomes via machine learning. In some models, the predicted variation effect on protein folding has a lower impact on severity than the predicted effect on protein function. This indicates that overall the correct folding of DDX3X might influence fetal survival, where we do not see severe outcomes in this case, which might apply to male patients in our study, whereas the functional impairment could be more important for the clinical outcome, which might apply to male and female patients in our study. Due to the sparse data used for training the machine learning model, the model predictions should not guide health-related behavior and/or clinical decisions.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003eAll data are within the paper and its Supporting Information files.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u0026nbsp;\u003c/strong\u003eWe thank the parents and caregivers of all included patients as well as the medical and para-medical personnel involved in patients\u0026rsquo; medical care.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003eThe study was concepted byEmanuele G. Coci. Patients\u0026rsquo; clinical and genetic data were acquired by Emanuele G. Coci, Maria Daniela D\u0026acute;Agostino, Laura Russell, Mouna Benamor, Alberto Fern\u0026aacute;ndez-Ja\u0026eacute;n, Ana Jimenez de Domingo, Lenka Noskova, Martin Magner, Sumit Parikh, Kimberly A Chapman, Himanshu Goel, Enrico Bertini, Ginevra Zanni, Ayman W. El-Hattab, Arif Khan, David Amor, A. Micheil Innes, Scott McLeod, Vanesa L\u0026oacute;pez-Gonz\u0026aacute;lez, Maria J. Ballesta-Martinez, Nino Spataro, Carmen Manso, Maria Piccione, Emanuela Salzano, Myriam Srour, Sarah Alsubhi, Deborah Renaud, Martina Baethmann, Matias Wagner, Pablo Prieto Matos, Jill A. Rosenfeld, Daryl A. Scott, Syed A. Ahmed, Kristyn L. Rawson, Fr\u0026oacute;\u0026eth;i Joensen, Marine Tessarech, Clement Proteau, Hong Li, David Michelson, Farzad Hashemi-Gorji, Audrey Putoux, Nicolas Chatron, Fanny Laffargue, Isabelle Creveaux, Rami Abou Jamra,\u0026nbsp;Virginia Pironi, Hilde Van Esch, Tania Cruz Marino, Jessica Ricard, Trine Bj\u0026oslash;rg Hammer, Irene Valenzuela, Amaia Lasa-Aranzasti, Anna M. Cueto-Gonzalez\u003cem\u003e,\u0026nbsp;\u003c/em\u003ePurificaci\u0026oacute;n Mar\u0026iacute;n Reina, Francisco Martinez\u003cem\u003e,\u0026nbsp;\u003c/em\u003eMilena Mariani\u003cem\u003e,\u0026nbsp;\u003c/em\u003eClaudia Ciaccio, Stefano D\u0026rsquo;arrigo\u003cem\u003e,\u0026nbsp;\u003c/em\u003eAnnalaura Torella, Vincenzo Nigro, Manuela Morleo, Elsebet Ostergaard, Maria Palomares-Bralo andFernando Santos-Simarro. Statistical calculations were performed by Pilar Chacon Millan. DDX3X monomer modellingwas performed by Christoph G.W. Gertzen and Holger Gohlke.\u003c/p\u003e\n\u003cp\u003eThe first draft of the manuscript was written by Emanuele G. Coci, Christoph G.W. Gertzen and Holger Gohlke; all other authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e The Center for Structural Studies is funded by the DFG (Grant number 417919780). We are grateful for computational support and infrastructure provided by the \u0026ldquo;Zentrum f\u0026uuml;r Informations- und Medientechnologie\u0026rdquo; (ZIM) at the Heinrich Heine University D\u0026uuml;sseldorf and the computing time provided by the John von Neumann Institute for Computing (NIC) to HG on the supercomputer JUWELS at J\u0026uuml;lich Supercomputing Centre (JSC) (user ID: VSK33, ETR1, DNAzyme).\u003c/p\u003e\n\u003cp\u003eMultiple authors are members of the European Reference Network on Rare Congenital Malformations and Rare Intellectual Disability ERN-ITHACA [EU Framework Partnership Agreement ID: 3HP-HP-FPA ERN-01-2016/739516].\u003c/p\u003e\n\u003cp\u003eThe work was supported by the project MULTIOMICS_CZ (Programme Johannes Amos Comenius, Ministry of Education, Youth and Sports of the Czech Republic,//ID Project CZ.02.01.01/00/23_020/0008540) \u0026ndash; Co-funded by the European Union and by grant NU23-07-00281 from the Ministry of Health of the Czech Republic. Instrumental support was provided by The National Center for Medical Genomics (LM2023067).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics statement\u0026nbsp;\u003c/strong\u003eParents or caregivers signed informed consent for publication of patients\u0026rsquo; anonymized medical records. The study was approved by the Danish Patients Safety Authority (protocol number: 3-3013-2462/1 and 3-3013-3205/1) and the Capital Region of Denmark (protocol Rh-2018-20-6158).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u0026nbsp;\u003c/strong\u003eThe authors have nofinancial or non-financial interest to disclose.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eYedavalli VS, Neuveut C, Chi YH, Kleiman L, Jeang KT (2004) Requirement of DDX3 DEAD box RNA helicase for HIV-1 Rev-RRE export function. Cell 119:381\u0026ndash;392\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVon Mueffling A, Garcia-Forn M, De Rubeis S (2024) DDX3X syndrome: From clinical phenotypes to biological insights. J Neurochem 168(9):2147\u0026ndash;2154\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSnijders Blok L, Madsen E, Juusola J, Gilissen C, Baralle D, Reijnders MR, Venselaar H, Helsmoortel C, Cho MT, Hoischen A et al (2015) Mutations in \u003cem\u003eDDX3X\u003c/em\u003e Are a Common Cause of Unexplained Intellectual Disability with Gender-Specific Effects on Wnt Signaling. Am J Hum Genet 97(2):343\u0026ndash;352\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang X, Posey JE, Rosenfeld JA, Bacino CA, Scaglia F, Immken L, Harris JM, Hickey SE, Mosher TM, Slavotinek A et al (2018) Phenotypic expansion in \u003cem\u003eDDX3X\u003c/em\u003e \u0026ndash; a common cause of intellectual disability in females. Ann Clin Transl Neurol 5(10):1277\u0026ndash;1285\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLennox AL, Hoye ML, Jiang R, Johnson-Kerner BL, Suit LA, Venkataramanan S, Sheehan CJ, Alsina FC, Fregeau B, Aldinger KA et al (2020) Pathogenic \u003cem\u003eDDX3X\u003c/em\u003e mutations impair RNA metabolism and neurogenesis during fetal cortical development. Neuron 106(3):404\u0026ndash;420\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eParra A, Pascual P, Cazalla M, Arias P, Gallego-Zazo N, San-Martin EA, Silv\u0026aacute;n C, Santos-Simarro F, Spanish OverGrowth Registry Initiative (SOGRI), Nevado J et al (2024) Genetic and phenotypic findings in 34 Spanish patients with \u003cem\u003eDDX3X\u003c/em\u003e neurodevelopmental disorder. Clin Genet 105(2):140-9\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKellaris G, Khan K, Baig SM, Tsai IC, Zamora FM, Ruggieri P, Natowicz MR, Katsanis N (2018) A hypomorphic inherited pathogenic variant in \u003cem\u003eDDX3X\u003c/em\u003e causes male intellectual disability with additional neurodevelopmental and neurodegenerative features. Hum Genomics 12(1):11\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBeal B, Hayes I, McGaughran J, Amor DJ, Miteff C, Jackson V, van Reyk O, Subramanian G, Hildebrand MS, Morgan AT et al (2019) Expansion of phenotype of DDX3X syndrome: six new cases. Clin Dysmorphol 28(4):169\u0026ndash;174\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNicola P, Blackburn PR, Rasmussen KJ, Bertsch NL, Klee EW, Hasadsri L, Pichurin PN, Rankin J, Raymond FL, Study DDD et al (2019) De novo \u003cem\u003eDDX3X\u003c/em\u003e missense variants in males appear viable and contribute to syndromic intellectual disability. Am J Med Genet A 179(4):570\u0026ndash;578\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eScala M, Torella A, Severino M, Morana G, Castello R, Accogli A, Verrico A, Vari MS, Cappuccio G, Pinelli M et al (2019) Three de novo \u003cem\u003eDDX3X\u003c/em\u003e variants associated with distinctive brain developmental abnormalities and brain tumors in intellectually disabled females. Eur J Hum Genet 27(8):1254\u0026ndash;1259\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDikow N, Granzow M, Graul-Neumann LM, Karch S, Hinderhofer K, Paramasivam N, Behl LJ, Kaufmann L, Fischer C, Evers C et al (2017) \u003cem\u003eDDX3X\u003c/em\u003e mutations in two girls with a phenotype overlapping Toriello-Carey syndrome. Am J Med Genet A 173(5):1369\u0026ndash;1373\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOkano S, Miyamoto A, Makita Y, Taketazu G, Kimura K, Fukuda I, Tanaka H, Yanagi K, Kaname T (2020) Severe gastrointestinal symptoms caused by a novel \u003cem\u003eDDX3X\u003c/em\u003e variant. Eur J Med Genet 63(12):104058\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen Y, Liu KY, Yang ZL, Li XH, Xu R, Zhou H (2020) A de novo \u003cem\u003eDDX3X\u003c/em\u003e variant is associated with syndromic intellectual disability: case report and literature review. Front Pediatr 8:303\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHu L, Xin X, Lin S, Luo M, Chen J, Qiu H, Ma L, Huang J (2020) A child with a novel \u003cem\u003eDDX3X\u003c/em\u003e variant mimicking cerebral palsy: a case report. Ital J Pediatr 46(1):88\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoresco G, Costanza J, Santaniello C, Rondinone O, Grilli F, Prada E, Orcesi S, Coro I, Pichiecchio A, Marchisio P et al (2021) A novel de novo \u003cem\u003eDDX3X\u003c/em\u003e missense variant in a female with brachycephaly and intellectual disability: a case report. Ital J Pediatr 47(1):81\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFieremans N, Van Esch H, Holvoet M, Van Goethem G, Devriendt K, Rosello M, Mayo S, Martinez F, Jhangiani S, Muzny DM et al (2016) Identification of Intellectual Disability Genes in Female Patients with a Skewed X-Inactivation Pattern. Hum Mutat 37(8):804\u0026ndash;811\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSobreira N, Schiettecatte F, Valle D, Hamosh A (2015) GeneMatcher: A Matching Tool for Connecting Investigators with an Interest in the Same Gene. Hum Mutat 36(10):928\u0026ndash;930\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSong H, Ji X (2019) The mechanism of RNA duplex recognition and unwinding by DEAD-box helicase DDX3X. Nat Commun 10(1):1\u0026ndash;8\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBas DC, Rogers DM, Jensen JH et al (2008) Very fast prediction and rationalization of pK(a) values for protein-ligand complexes. Proteins 73(3):765\u0026ndash;783\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCase DA, Metin Aktulnga H, Belfon K, Ben-Shalom I, Berryman JT, Brozell SR, Cerrutti DS, Cheatham TE, Cisneros GA, Cruzeiro VWD et al (2022) Amber22; University of California, San Francisco\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang J, Cieplak P, Kollman P (2000) How well does a restrained electrostatic potential (RESP) model perform in calculating conformational energies of organic and biological molecules? J Comput Chem 21(12):1049\u0026ndash;1074\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZgarbova M, Otyepka M, Sponer J, Mladek A, Banas P, Cheatham TE, Jurecka P (2011) Refinement of the Cornell et al. nucleic acids force field based on reference quantum chemical calculations of glycosidic torsion profiles. J Chem Theory Comput 7(9):2886\u0026ndash;2902\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBanas P, Hollas D, Zgarbova M, Jurecka P, Orozco M, Cheatham TE, Sponer J, Otyepka M (2010) Performance of molecular mechanics force fields for RNA simulations: stability of UUCG and GNRA hairpins. J Chem Theory Comput 6(12):3836\u0026ndash;3849\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerez A, Marchan I, Svozil D, Sponer J, Cheatham TE, Laughton C, Oroczo M (2007) Refinement of the AMBER force field for nucleic acids: improving the description of α/γ conformers. Biophys J 92(11):3817\u0026ndash;3829\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHumphrey W, Dalke A, Schulten K (1996) VMD: visual molecular dynamics. J Mol Graphics 14(1):33\u0026ndash;38\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJumper J, Evans R, Prityel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool K, Bates R, Zidek A, Potapenko A et al (2021) Highly accurate protein structure prediction with AlphaFold. Nature 596(7873):583\u0026ndash;589\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDelgado J, Radusky LG, Cianferoni D, Serrano L (2019) FoldX 5.0: working with RNA, small molecules and a new graphical interface. Bioinformatics 35(20):4168\u0026ndash;4169\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMann H, Whitney D (1947) Controlling the false discovery rate: A practical and powerful approach to multiple testing. Ann Math Stat 18(1):50\u0026ndash;60\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWilcoxon F (1992) Individual comparisons by ranking methods. In: Kotz S, Johnson NL (eds) Breakthroughs in Statistics. Springer, New York, pp 196\u0026ndash;202\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDai Y, Yang Z, Guo J, Li H, Gong J, Xie Y, Xiao B, Wang H, Long L (2022) Expansion of Clinical and Genetic Spectrum of \u003cem\u003eDDX3X\u003c/em\u003e Neurodevelopmental Disorder in 23 Chinese Patients. Front Mol Neurosci 15:793001\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJayaraman K, Trachtmann N, Sprenger GA, Gohlke H (2022) Protein engineering for feedback resistance in 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase. Appl Microbiol Biotechnol 106(19\u0026ndash;20):6505\u0026ndash;6517\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBabu MM (2016) The contribution of intrinsically disordered regions to protein function, cellular complexity, and human disease. Biochem Soc T 44(5):1185\u0026ndash;1200\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJemth P, Karlsson E, Vogeli B, Guzovsky B, Andersson E, Hulquist G, Dogan J, Guntert P, Riek R, Chi CN (2018) Structure and dynamics conspire in the evolution of affinity between intrinsically disordered proteins. Sci Adv 4(10):eaau4130\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoulat D, Buerckstuemmer T, Westermayer S, Goncalves A, Bauch A, Stefanovic A, Hantschel O, Bennett KL, Decker T, Superti-Furga G (2008) The DEAD-box helicase DDX3X is a critical component of the TANK‐binding kinase 1‐dependent innate immune response. EMBO J 27(15):2135\u0026ndash;2146\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 and 2 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"DDX3X associated multisystem disorder, DDX3X variants’ type and position, DDX3X genotype-phenotype correlation, In-silico simulation of mutatnt DDX3X monomers","lastPublishedDoi":"10.21203/rs.3.rs-7842722/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7842722/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cem\u003eDDX3X\u003c/em\u003e dysfunction causes an X-linked multisystem disorder with high penetrance and variable expressivity. The phenotypic spectrum spans from learning disability without somatic involvement to profound intellectual disability with severe impairments in the central nervous system and other organs. A few multicenter studies and single case reports have previously highlighted some common phenotypic patterns but were unable to delineate correlations between underlying \u003cem\u003eDDX3X\u003c/em\u003e variants and phenotypic findings.\u003c/p\u003e\u003cp\u003eFrom the second largest patient cohort published to date, we analysed clinical, psychometric and diagnostic findings of 52 female and 7 male individuals, harbouring \u003cem\u003ede novo\u003c/em\u003e and inherited \u003cem\u003eDDX3X\u003c/em\u003e variants. These female patients revealed previously unknown quantitative correlations between variant type and localization and specific phenotypic findings (growth features, epilepsy, brain anomalies, dysmorphisms, motor-focused neurological findings). Moreover, by analysing the \u003cem\u003ein silico\u003c/em\u003e folding and RNA binding capability of mutant DDX3X monomers, we were able to delineate novel correlations between DDX3X monomer misfolding grade and phenotypic severity.\u003c/p\u003e","manuscriptTitle":"DDX3X syndrome: a multicenter genotype-phenotype study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-11 16:19:26","doi":"10.21203/rs.3.rs-7842722/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c2460955-f8ee-4007-9f9c-19c315ff878a","owner":[],"postedDate":"November 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-26T20:43:54+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-11 16:19:26","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7842722","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7842722","identity":"rs-7842722","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}