Copy-number variants as modulators of common disease susceptibility

21

Copy-number variations (CNVs) have been associated with rare and 22 debilitating genomic syndromes but their impact on health later in life in the general population 23 remains poorly described. 24

Assessing four modes of CNV action, we performed genome-wide association 25 scans (GWASs) between the copy-number of CNV -proxy probes and 60 curated ICD-10 26 based clinical diagnoses in 331,522 unrelated white UK Biobank participants with replication 27 in the Estonian Biobank. 28

We identified 73 signals involving 40 diseases, all of which indicating that CNVs 29 increased disease risk and caused earlier onset. Even after correcting for these signals, a 30 higher CNV burden increased risk for 18 disorders, mainly through the number of deleted 31 genes, suggesting a polygenic CNV architecture. Number and identity of genes disturbed by 32 CNVs affected their pathogenicity, with many associations being supported by colocalization 33 with both common and rare single nucleotide variant association signals . Dissection of 34 association signals provided insights into the epidemiology of known gene-disease pairs (e.g., 35 deletions in BRCA1 and LDLR increased risk for ovarian cancer and ischemic heart disease, 36 respectively), clarified dosage mechanisms of action (e.g., both increased and decreased 37 dosage of 17q12 impacts renal health), and identified putative causal genes (e.g., ABCC6 for 38 kidney stones). Characterization of the pleiotropic pathological consequences of recurrent 39 CNVs at 15q13, 16p13.11, 16p12.2, and 22q11.2 in adulthood indicated variable expressivity 40 of these regions and the involvement of multiple genes. 41

Our results shed light on the prominent role of CNVs in determining common 42 disease susceptibility within the general population and provide actionable insights allowing 43 to anticipate later-onset comorbidities in carriers of recurrent CNVs. 44 45

46 Structural variation; CNV; GWAS; common diseases; pleiotropy; genomic disorders. 47 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 3

48 Copy-number variants (CNVs) refer to duplicated or deleted DNA fragments (≥ 50bp) and 49 represent an important source of inter-individual variation [1,2]. As a highly diverse mutational 50 class, they can alter the copy-number of dosage sensitive genes, induce gain- or loss-of-51 function (LoF) through gene fusion or truncation, unmask recessive alleles, or disrupt 52 regulatory sequences, thereby representing potent phenotypic modifiers [3]. As such, t heir 53 role in human disease has mainly been studied in clinically ascertained cohorts often 54 presenting with congenital anomalies and/or severe neurological (e.g., developmental delay 55 and intellectual disability, epilepsy) or psychiatric (e.g., autism or schizophrenia) symptoms 56 [4–7] and today, close to 100 genomic disorders (i.e., disease caused by genomic 57 rearrangements) have been described [8,9]. D espite their deleteriousness, some of these 58 CNVs flanked by repeats recurrently appear and remain at a low but stable frequency in the 59 population [10]. 60 61 The emergence of large biobanks coupling genotype to phenotype data has fostered the study 62 of CNVs in the general population. Whole genome sequencing represents the best approach 63 to characterize the full human CNV landscape [1,11,12] but current long- and short -read 64 sequencing association studies have limited samples size [13–15]. Alternatively, l arger 65 sample sizes are available for exome sequencing data, offering the possibility to assess the 66 phenotypic consequence of small CNVs [16,17], while microarray-based CNV calls are better-67 suited for the study of large CNV s and have been successful ly used in association studies 68 [8,18–28]. Performing a CNV genome -wide association study (GWAS) on 57 medically 69 relevant continuous traits in the UK Biobank (UKBB) [29], we previously identified 131 70 independent associations, including allelic series wherein carriers of CNVs at loci previously 71 associated with rare Mendelian disorders exhibited subtle changes in disease-associated 72 phenotypes but lacked the corresponding clinical diagnosis [23]. Paralleling findings for point 73 mutations [30–33], this supports a model of variable expressivity, where CNVs can cause a 74 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 4 wide spectrum of phenotypic alteration ranging from severe, early -onset diseases to mild 75 subclinical symptoms, opening the question as to whether these loci are also associated with 76 common diseases. 77 78 Unlike continuous traits that can be objectively measured in all participants , population 79 cohorts, such as UKBB, have low numbers of diseased individuals [34]. Moreover, defining 80 cases relies on the arbitrary dichotomization of complex underlying pathophysiological 81 processes [35]. Beyond the inherent loss of power associated to usage of binary variables 82 [36], cases might be missed because an individual did not consult a physician, was 83 misdiagnosed due to atypical clinical presentation, or is in a prodromal disease phase. Studies 84 investigating CNV-disease associations in the general population have either focused on only 85 few diseases [27,37] or well-established recurrent CNVs [20,38,39]. Alternatively, high-86 throughput studies have assessed a broad range of continuous and binary traits 87 simultaneously [17,24,25] without any precautions to accommodate the aforementioned 88 challenges. To date, the largest disease CNV-GWAS meta-analyzed ~1,000,000 individuals 89 [8]. While boost ing power through increased sample size, it comes at the cost of extensive 90 data harmonization, resulting in the exclusion of smaller CNVs (≤ 100kb) and broader disease 91 categories (e.g., “immune a bnormality”). Moreover, as th is study includes several clinical 92 cohorts, phenotypes are biased towards neuropsychiatric disorders (24/54 phenotypes) for 93 which the role of CNVs is well-established [4–7]. 94 95 Using tailored CNV -GWAS models mimicking four mechanisms of CNV action and time -to-96 event analysis , we investigate the relation ship between CNVs and 60 carefully defined 97 common diseases affecting a broad range of physiological systems in 331,522 unrelated white 98 UKBB participants. Extensively validating our results , we report associations according to 99 confidence tiers and take advantage of rich individual-level phenotypic data to demonstrate 100 the contribution of CNVs to the common disease burden in the general population. 101 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 5

102 1. Study material 103 Discovery cohort: UK Biobank 104 The UK Biobank (UKBB) is composed of ~500,000 volunteers (54% females) from the general 105 UK population for which microarray -based genotyping and extensive phenotyping data – 106 including hospital based International Classification of Diseases, 10th Revision (ICD -10) 107 codes (up to September 2021) and self-reported conditions – are available [29]. Participants 108 signed a broad informed consent form and data were accessed through application #16389. 109 110 Replication cohort: Estonian Biobank 111 The Estonian Biobank (EstBB) is a population-based cohort of ~208,000 Estonian individuals 112 (65% females; data freeze 2022v01 [12/04/2022]) for which microarray-based genotyping data 113 and ICD-10 codes from crosslinking with national and hospital databases (up to end 2021) are 114 available [40]. The activities of the EstBB are regulated by the Human Genes Research Act, 115 which was adopted in 2000 specifically for the operations of the EstBB. Individual level data 116 analysis in the EstBB was carried out under ethical approval 1.1 -12/624 from the Estoni an 117 Committee on Bioethics and Human Research (Estonian Ministry of Social Affairs), using data 118 according to release application 3-10/GI/34668 [20/12/2022] from the EstBB. All participants 119 signed a broad informed consent form. 120 121 Other public resources 122 The PheCode Map 1.2 (beta) (https://phewascatalog.org/phecodes_icd10) was used for ICD-123 10 code classification [41]. Genomic regions were annotated with the NHGRI-EBI GWAS 124 Catalog (https://www.ebi.ac.uk/gwas/; 24/11/2022) [42], and the Online Mendelian Inheritance 125 in Man (OMIM; https://www.omim.org/; 27/07/2022) [43]. Recurrent CNV coordinates w ere 126 retrieved from DECIPHER ( https://www.deciphergenomics.org/) [9]. Unless specified 127 otherwise, Neale UK Biobank single-nucleotide polymorphism (SNP)-GWASs summary 128 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 6 statistics were used ( http://www.nealelab.is/uk-biobank). Allele frequencies and genomic 129 constraint scores (probability of LoF Intolerance (pLI); LoF Observed over Expected Upper 130 bound Fraction (LOEUF) ) originate from the Genome Aggregation Database (GnomAD; 131 https://gnomad.broadinstitute.org/) [44]; pHaplo and pTriplo scores from [8]. Tissue-specific 132 gene expression was assessed in the Genotype -Tissue Expression project (GTEx; 133 https://gtexportal.org/home/) [45]. 134 135 Software versions 136 CNVs were called with PennCNV v1.0.5 [46] using PennCNV-Affy (27/08/2009) and filtered 137 based on a quality scoring pipeline [47]. Genetic analyses were conducted with PLINK v1.9 138 and v2.0 [48]. ANNOVAR (24/10/2019) was used to map genes to genetic regions [49]. The 139 UCSC Genome Browser was used to determine the human genome size (GRCh37/hg19) and 140 the LiftOver tool was used to lift over genomic coordinates [50]. Statistical analyses were 141 performed with R v3.6.1 and graphs were generated with R v4.1.3. 142 143 2. CNV association studies in the UK Biobank 144 Microarray-based CNV calling 145 UKBB g enotype microarray data were acquired from two arrays with 95% probe overlap 146 (Applied Biosystems UK Biobank Axiom Array : 438,427 samples; Applied Biosystems UK 147 BiLEVE Axiom Array by Affymetrix: 49,950 samples) [29] and used to call CNVs as previously 148 described [23]. Briefly, CNVs were called using standard PennCNV settings and samples on 149 genotyping plates with a mean CNV count per sample > 100 and samples with > 200 CNVs 150 or single CNV > 10 Mb were excluded. Remaining CNVs were attributed a probabilistic quality 151 score (QS) ranging from -1 (likely deletion) to 1 (likely duplication) [47]. High confidence CNVs, 152 stringently defined by |QS| > 0.5, were retained and encoded in chromosome-wide probe-by-153 sample matrices (i.e., entries of 1, -1, or 0 indicate probes overlapping high confidence 154 duplication, deletion, or no/low quality CNV, respectively) [47], which were converted into three 155 PLINK binary file sets to accommodate association analysis according to four modes of CNV 156 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 7 action. Details about the CNV encoding and handling of chromosome X are provided in 157 Supplemental Note 1. Probe-level CNV frequency was calculated [23]. All results in this study 158 are based on the human genome reference build GRCh37/hg19. 159 160 Case-control definition and age-at-disease onset calculation 161 A pool of 331,522 unrelated white British UKBB participants (54% females) was considered 162 after excluding retracted (up to August 2020), as well as related, high missingness and non-163 white British samples (used.in.pca.calculation = 0 and in.white.British.ancestry.subset = 0 in 164 Sample-QC v2 file) . CNV outliers (Microarray-based CNV calling) and individuals reporting 165 blood malignancies (i.e., possibly harboring somatic CNVs; UKBB field #20001: 10047, 1048, 166 1050, 1051, 1052, 1053, 1055, 1056, 1056 ; #41270: ICD-10 codes mapping to PheCode 167 exclusion range “cancer of lymphatic and hematopoietic tissue”), were further excluded. 168 169 Cases and controls were assigned for 60 ICD-10-based clinical diagnoses using diagnosis – 170 ICD10 (#41270), cancer code, self -reported (#20001), and non-cancer illness code, self -171 reported (#20002) to build exclusion and inclusion lists. For each disease, we first defined all 172 331,522 individuals as control s. We excluded individuals with a self-reported or hospital 173 diagnosis of a broad set of conditions that include the disease of interest, as well as related 174 disorders/ICD-10 codes (e.g., other cancers or radio-/chemotherapy for breast cancer; mood 175 or personality disorders for schizophrenia). We then re-introduce as cases individuals having 176 received a restricted range of ICD-10 codes matching our disease definition. For second level 177 ICD-10 codes, all subcodes are considered, otherwise only the specified ones. The disease 178 burden was calculated as the number of diagnoses (out of the 60 assessed) an individual has 179 received. For male- (prostate cancer) and female - (menstruation disorders, endometriosis, 180 breast cancer, ovarian cancer ) specific diseases, downstream analyses were conducted 181 excluding individuals from the opposite sex. 182 183 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 8 Based on the date at first in -patient diagnosis – ICD10 (#41280) and the individual’s month 184 (#52) and year (#34) of birth (birthday assumed on average to be the 15th), the age at diagnosis 185 was calculated by subtracting the earlie st diagnosis date for codes on the inclusion list from 186 the birth date and converting it to years by dividing by 365.25 to account for leap years. 187 188 Probe and covariate selection 189 Relevant covariates and probes were pre-selected to fit tailored main CNV-GWAS models and 190 reduce computation time. For each disease, a logistic regression was fitted to explain disease 191 probability as a function of age (#21003), sex, genotyping array , and the 40 first principal 192 components (PCs). Nominally significantly associated covariates (p ≤ 0.05) were retained for 193 the main analysis . CNV-proxy probes with a CNV frequency ≥ 0.01% were pruned at r2 > 194 0.9999 in PLINKCNV (–-indep-pairwise 500 250 0.9999 PLINK v2.0) to group probes 195 at the core of CNV regions while retaining resolution at breakpoints (BPs), resulting in 18,725 196 probes. For each disease , 2-by-3 genotypic Fisher test s assessed dependence between 197 disease status and probe copy-number (rows: control versus case; columns: deletion versus 198 copy-neutral versus duplication; –-model fisher PLINK v1.9; TEST column GENO). Probes 199 with p ≤ 0.001 and a minimum of two disease cases among CNV, duplication, or deletion 200 carriers were retained for assessment through the mirror/U -shaped, duplication -only, or 201 deletion-only model, respectively. 202 203 Genome-wide significance threshold 204 Due to the recurrent nature of CNVs, the 18,725 probes retained after frequency filter and 205 pruning remain highly correlated and are thus not independent. Accounting for all of them 206 would result in an overly strict multiple testing correction. Using an established protocol 207 [22,23,51], we estimated the number of effective tests performed to Neff = 6,633, setting the 208 genome-wide (GW) threshold for significance at p ≤ 0.05/6,633 = 7.5 x 10-6. This threshold is 209 of the same order of magnitude as what others have estimated for disease CNV-GWAS [8]. 210 211 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 9 Main CNV-GWAS model 212 Association between disease risk and copy-number of CNV-proxy probes was assessed 213 through logistic regression with Firth fallback (--covar-variance-standardize –-glm 214 firth-fallback omit-ref no-x-sex hide-covar -–ci 0.95 PLINK v2.0), using 215 disease- and model -specific probes and covariates ( Probe and covariate selection ). Four 216 association models were assessed: the mirror model assesses the additive effect of each 217 additional copy (PLINKCNV); the U-shape model assesses a consistent effect of any deviation 218 from the copy neutral state (PLINKCNV, using the hetonly option in glm PLINK v2.0); the 219 duplication-only model (PLINKDUP) assesses the impact of a duplication while disregarding 220 deletions; the deletion-only model (PLINKDEL) assesses the impact of a deletion while 221 disregarding duplications. Odds ratios (OR) and their 95% confidence interval (CI) were 222 harmonized (A1 to “T”; Supplemental Note 1 – Table 1 ), i.e., 𝑂𝑅!"# = $ %&!"# and 223 𝐶𝐼!"# = 𝑒'() (%&$%&)±$../∗12'() (,-!"#), respectively. GW-significant associations (p ≤ 7.5 x 10-6; 224 Genome-wide significance threshold ) were pruned at r2 > 0.8 (--indep-pairwise 3000 225 500 0. 8 PLINK v2 .0), giving priority to probes with the strongest association signal by 226 inputting a scaled negative logarithm of association p -value as frequency (–-read-freq 227 PLINK v2.0). For the U-shape model, pruning was performed using custom code by extracting 228 probes from PLINKCNV and recoding them to match U-shape numerical encoding. Number of 229 independent signals per disease was determined by stepwise conditional analysis. Briefly, for 230 each disease and association model, the numerical CNV genotype of the lead probe was 231 included along selected covariates in the logistic regression model. This process was repeated 232 until no more GW-significant signal remained. 233 234 Due to its continuous nature , the disease burden CNV -GWAS was based on linear 235 regressions between copy-number of selected probes and the disease burden (–-glm omit-236 ref no-x-sex hide-covar allow-covars PLINK v2 .0), correcting for selected 237 covariates. Post-GWAS processing was performed as previously described [23]. 238 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 10 239 CNV region definition and annotation 240 CNV region (CNVR) boundaries were defined by the most distant probe within ± 3Mb and r2 ≥ 241 0.5 of independent lead probes (–-show-tags --tag-kb 3000 --tag-r2 0.5 PLINK 242 v1.9; U -shape model: custom code, as described previously for pruning). Signals from 243 different models were merged when overlapping (≥ 1bp) and involving the same disease, with 244 CNVR boundaries defined as the maximal CNVR. Characteristics of the most significant 245 model (i.e., “best model”) are reported. The “main model” indicates which CNV type mainly 246 drives the association, i.e., when associations were found through multiple models, priority 247 was given to either the duplication-only or deletion -only models, otherwise to the model 248 yielding the lowest p -value. CNVRs were annotated with hg19 HGNC and ENSEMBL gene 249 names using annotate_variation.pl from ANNOVAR (--geneanno). Number of genes 250 mapped to a CNVR was calculated and set to zero for CNVRs with REGION not equaling 251 “exonic”. Resulting count was used as a predictor for pleiotropy (i.e., number of associations) 252 through linear regression. 253 254 Statistical confidence tiers 255 Following primary assessment through logistic regression (Main CNV-GWAS model), three 256 statistical approaches were implemented to gauge robustness of the lead probe’s association 257 signal. First, we assessed post hoc the p-value of 2-by-3 genotypic Fisher tests (Probe and 258 covariate selection). Second, we transformed the binary disease status into a continuous 259 variable by computing the response residuals of the logistic regression of disease status on 260 disease-relevant covariates. This allowed the usage of linear regressions to estimate the effect 261 of the CNV genotype (encoded according to all significantly associated models in the primary 262 analysis) on disease risk. The model generating the lowest p-value for the CNV encoding is 263 reported. Third, time-to-event analysis was used to assess whether CNVs influence age -at-264 disease onset. Age at last healthy measurement was calculated as age -at-disease onset for 265 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 11 cases and date of last recorded diagnosis (30/09/2021) minus birth date converted to years 266 for controls (Case-control definition and age-at-disease onset calculation). Cox proportional-267 hazards (CoxPH) models were fitted including disease -relevant covariates and numerically 268 encoded CNV genotype for either of the four association models as predictors, using coxph() 269 function from the R survival package [52]. The model with the lowest CNV genotype p-270 value is reported. CNV-disease associations were classified in confidence tiers depending on 271 whether they were confirmed by 3 (tier 1) , 2 (tier 2) , or 1 (tier 3) of the above -described 272 approaches at the arbitrary validation significance threshold of p ≤ 1 x 10-4. Validation 273 approaches not being suited for continuous variables – which do not suffer from the same 274 caveats as binary traits – all disease burden associations were classified as tier 1. 275 276 Literature-based supporting evidence 277 Using three literature-based approaches, we examined whether disease-associated CNVRs 278 had previously been linked to relevant phenotypes. First, we investigated the colocalization of 279 autosomal CNVRs with SNP-GWAS signals. GRCh38/hg38 lifted CNVR coordinates were 280 inputted in the GWAS Catalog and associations (p ≤ 1x10 -7) relevant to the inve stigated 281 disease (i.e., synonym, continuous proxy, or major risk factor) were identified through manual 282 curation. Second, we overlapped OMIM morbid genes (i.e., linked to an OMIM disorder; 283 morbidmap.txt) with disease-associated CNVRs. Through manual curati on, we flagged 284 OMIM genes associated to Mendelian disorders sharing clinical features with the common 285 disease associated through CNV-GWAS. Third, we examined if implicated CNVRs overlapped 286 regions at which CNVs were found to modulate continuous traits [23] or disease risk [20,25]. 287 288 3. Replication in the Estonian Biobank 289 CNV calling and sample selection 290 Autosomal CNVs were called from Illumina Global Screening Array (GSA) genotype data for 291 193,844 individuals that survived general quality control and had matching genotype -292 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 12 phenotype identifiers, matching inferred versus reported sex, a SNP-call rate ≥ 98%, and were 293 included in the EstBB SNP imputation pipeline. CNV outliers and individuals with a reported 294 blood malignancy were excluded, as previously described. High confidence CNV calls (|QS| 295 > 0.5) of the 156,254 remaining individuals were encoded into three PLINK binary file sets, 296 following the procedure described for the UKBB (CNV association studies in the UK Biobank). 297 298 EstBB disease definition 299 Disease cases and disease burden were defined similarly than in the UKBB. To account for 300 differences in recording practices between the countries, Z12 (routine preventive screens for 301 cancer), and D22-23 (benign skin lesions) subcodes, were removed from the exclusion list of 302 cancer traits as they were much more frequent than in the UKBB and strongly reduced the 303 number of controls. Due to lack of matching data in the EstBB, no self-reported diseases and 304 cancers were used as an exclusion criterion for disease definition. 305 306 EstBB replication analysis 307 Related individuals with available CNV calls were pruned (KING kinship coefficient > 0.0884), 308 prioritizing individuals whose disease status was least often missing, leaving 90,211 unrelated 309 samples for the replication study. Disease-relevant covariates were selected among sex, year 310 of birth, genotyping batch (1 -11), and PC1 -20. For each of the 73 UKBB signals, probes 311 overlapping the CNVR and with a n EstBB CNV, duplication, or deletion frequency ≥ 0.01%, 312 were retained, depending on whether the mirror/U -shape, duplication-only, or deletion -only 313 was the best UKBB model , respectively. Association studies were performed on remaining 314 probes using disease-specific covariates and the best UKBB model, following the previously 315 described procedure. Forty signals (55%) could not be assessed due to null/low CNV 316 frequency, failure of the regression to converge, absence of at least one case CNV carrier, or 317 because not mapping on the autosomes. For the remaining 33 signals, summary statistics of 318 the probe showing the strongest association within the CNVR were retained and p-values 319 were adjusted to account for directional concordance with UKBB effect s by rewarding and 320 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 13 penalizing signals with matching and non-matching effect size signs, respectively. Specifically, 321 one-sided p-values were obtained as 𝑝!"# = 3!"# 4 and 𝑝!"# = 1 − ( 3!"# 4 ) for 24 concordant and 322 9 non -concordant signals, respectively. Accounting for 33 testable signals, the replication 323 threshold for significance was set at p ≤ 0.05/ 33 = 1.5 x 10-3. One-sided binomial tests 324 (binom.test()) were used to assess e nrichment of observed versus expected significant 325 associations at various thresholds ( 𝛼 = 0.1 to 0.005 by steps of 0.005) , with the R function 326 arguments: 𝑥 the number of observed signals at 𝛼, 𝑛 the number of testable signals (i.e., 33), 327 and 𝑝 the expected probability of signals meeting 𝛼 (i.e., 𝛼). 328 329 4. CNV region constraint analysis 330 Evolutionary c onstraint of genes overlapping disease-associated CNVRs , i.e., “disease 331 genes” (CNV region definition and annotation), was assessed by comparing their pLI, LOEUF, 332 pHaplo, and pTriplo scores to the ones of “background genes”. The latter were identified by 333 annotating ranges of one or multiple consecutive probes with CNV frequency ≥ 0.01% with 334 ANNOVAR (hg19 HGNC gene names) and excluding disease genes. For pLi and LOEUF, all 335 disease genes were considered together. For pHaplo and pTriplo, two disease gene groups 336 were considered: genes overlappin g CNVRs with at least one association through the 337 duplication-only model and genes overlapping CNVRs with at least one association through 338 the deletion-only model. As many CNVRs associated through both models, the analysis was 339 repeated considering genes ov erlapping CNVRs with at least one association through the 340 duplication-only and none through the deletion -only model and vice-versa. Comparison with 341

genes was done through two-sided Wilcoxon rank-sum test. 342 343 5. Extended phenotypic assessment 344 To elaborate on specific associations, we made use of the rich phenotypic data available for 345 UKBB participants , as detailed in Supplemental Note 2. For fine -mapping of association 346 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 14 signals, CNV carriers were divided in subgroups based on visual inspection of C NV 347 breakpoints and segmental duplications, as detailed in Supplemental Note 3. 348 349 CNV versus copy-neutral comparisons 350 Comparisons between groups of CNV carriers and copy -neutral individuals always exclude 351 low quality CNV (|QS| ≤ 0.5) carriers altogether. For diseases, prevalence is estimated as 𝑞 =352 5 ! , with 𝑐 and 𝑛 are the number of cases and total number of individuals in a group, and 353 𝑆𝐸(𝑞) = 4 6∗($76) ! . Differences in prevalence compared to cop y-neutral individuals were 354 assessed by two-sided Fisher test. For continuous traits, comparisons are based on two-sided 355 t-tests. 356 357 6. CNV burden analyses 358 CNV burden association studies 359 In the UKBB, individual -level CNV, duplication, and deletion burden were calculated as the 360 number of Mb or genes affected by high -confidence (|QS| > 0.5) autosomal CNVs , 361 duplications, and deletions, respectively, as previously described [23]. Association between 362 burden values and the 60 diseases (logistic regress ion) or the di sease burden (linear 363 regression), was assessed including disease-relevant covariates in the model. Accounting for 364 the 61 evaluated traits, significance was defined at p ≤ 0.05/61 = 8.2 x 10-4. We next corrected 365 burden values for CNV-GWAS signals. For each disease, CNVs, duplications, and deletions 366 overlapping (≥ 1bp) a CNVR significantly associated with the disease of interest through CNV-367 GWAS were omitted from the CNV, duplication, and deletion burden calculations if the CNVR 368 had been found to associate with the disease through the mirror/U-shape, duplication-only, or 369 deletion-only model, respectively. Association studies were repeated using corrected burden 370 values. Only the most significant burden types are reported in the text. 371 372 Relative importance of protein coding regions in mediating the burden’s effect 373 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 15 The average genome -wide gene density ( 𝐺𝐷89) was estimated to 8.4 genes/Mb based on 374 26,289 genes (i.e., unique HGNC gene names in hg19 RefSeq, excluding microRNAs but 375 including genes of uncertain function ( i.e., “LOC”)) and a human genome length of 376 3,137,161,264 bp. If CNVs affecting the coding and non-coding DNA have similar effects, we 377 expect that the impact of 1Mb affected by CNVs to be equivalent to 8.4 genes being affected 378 by CNVs. Hence, the association effect size of the CNV burden measured in Mb (𝛽:;) is 379 expected to be 8.4-times larger than the one measured in number of affected genes (𝛽<"!"), 380 i.e., =/0 =1%$% =8.4. This hypothesis was tested independently for the deletion and duplication 381 burdens for traits with at least one significant uncorrected burden association (i.e., 20 diseases 382 + disease burden). Significant deviations from the expected ratio were assessed by t-statistic: 383 𝑡 = 8.4 − 𝛽<:; 𝛽<<"!" 𝑆𝐷= 384 where 𝛽<:; and 𝛽<<"!" are the estimated effects of the burden measured in Mb or number of 385 genes impacted by CNVs, respectively, on the assessed trait. 𝑆𝐷= is the empirically observed 386 standard deviation of the =>/0 =>1%$% ratio, estimated based on 10,000 simulations of 𝛽>:; ~ 𝑁(𝛽<:;,387 𝑉𝑎𝑟E (𝛽<"!" ~ 𝑁(𝛽<<"!" , 𝑉𝑎𝑟E (𝛽<<"!" )). P-values were computed based on a two-sided 388 one sample t -test and deemed significant at p ≤ 0.05/ 21 = 2.4 x 10-3. This analysis was 389 repeated for the modified burden definition s not accounting for CNVs overlapping disease -390 associated CNVRs. 391 392

393 The spectrum of common diseases in the UK Biobank 394 To mitigate issues related to disease definition, we used a three-step approach to designate 395 cases and controls in the UKBB (Figure 1A; top). Starting from 331,522 unrelated white British 396 individuals, we defined cases based on a narrow list of hospital -based diagnoses (i.e., ICD-397 10 codes) and excluded self-reported cases, as well as self-reported and hospital diagnoses 398 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 16 of related conditions. Sixty disorders spanning 12 ICD -10 chapters were selected to cover a 399 wide range of physiological systems, favoring conditions with sufficiently large sample size 400 and a likely genetic basis ( Figure 1B ; Figure S1A; Table S1 ). Except for systemic lupus 401 erythematosus (N = 422) and polycystic kidney disease (N = 454), all diseases had over 500 402 cases. Nineteen diseases had a case count > 10,000, with osteoarthrosis (N = 62,175) and 403 essential hypertension (N = 97,860) being the most frequent. Seven diseases h ad a median 404 age of onset ≤ 60 years , predominantly female reproductive disorders , autoimmune 405 conditions, and psychiatric disorders. Conversely, the nine diseases with a median age at 406 onset ≥ 70 years were mainly degenerative disorders of the brain, eye , and kidney, overall 407 aligning align with epidemiological knowledge of the respective diseases.408 409 Figure 1. Overview of the study 410 (A) Schematic representation of the analysis workflow. Trait definition: For each of the 60 investigated 411 diseases, unrelated white British UK Biobank participants were assigned as controls, individuals self -412 reporting or diagnosed with the disease of interest or a broader set of related conditions were excluded 413 and set as missing, and individuals with a hospital-based ICD-10 diagnosis of the condition of interest 414 were re-introduced as cases. Primary association study: Disease -specific relevant covariates were 415 selected. Probes were pre -filtered based on copy -number variant (CNV) frequency , required to 416 associate with the disease, and a minimum of two diseased carriers was required for the probe to be 417 carried forward. Disease- and model-specific covariates and probes were used to generate tailored 418 CNV genome-wide association studies (GWASs) based on Firth fallback logistic regression according 419 to a mirror, U -shape, duplication -only (i.e., considering only duplications), and deletion -only (i.e., 420 considering only deletions) models. Independent lead signals were identified through stepwise 421 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 17 conditional analysis and CNV regions were defined based on probe correlation and merged across 422 models. Validation: Statistical validation methods (i.e., Fisher test, residuals regression , and Cox 423 proportional hazards model (CoxPH)) were used to rank associations in confid ence tiers. Literature 424 validation approaches leverage data from independent studies to corroborate that genetic perturbation 425 (i.e., single-nucleotide polymorphisms (SNP), rare variants from the OMIM database, and CNVs) in the 426 region are linked to the disease. Independent replication in the Estonian Biobank. (B) Age at onset for 427 the 60 assessed diseases, categorized based on ICD-10 chapters and colored according to case count. 428 Data are represented as boxplots; outliers are not shown. 429 Copy-number variant genome-wide association study 430 To assess whether disease susceptibility is modulated by CNVs, we performed CNV genome-431 wide association studies (GWASs), i.e., test if the copy-number of selected probes influence 432 the probability to develop a disease or an individual’s disease burden (i.e., number of 433 diagnoses among the 60 studied diseases ) (see Methods; Figure 1A ; middle). Briefly, 434 microarray-called CNVs for 331,522 unrelated white British individuals were transformed to 435 the probe level after quality -control [23]. As CNVs can act through different gene dosage 436 mechanisms, four association models were assessed: mirror and U-shape models consider 437 deletions and duplications simultaneously, assuming that they impact disease risk in opposite 438 or identical direction, respectively, while the CNV type-specific duplication- and deletion-only 439 models assess independently the effect of duplications and deletions, respectively. To reduce 440 the number and complexity of implemented logistic regressions, pre-processing steps 441 selected relevant covariates and probes for each disease and model combination , thereby 442 lowering computation time and decreasing the multiple testing burden (Figure S2; Table S2). 443 All summary statistics are available (Data Availability). 444 445 Stepwise conditional analysis narrowed GW significant associations (p ≤ 7.5 x 10 -6; see 446

for threshold calculation) to 40, 41, 21, and 38 independent signals for the mirror, U-447 shape, duplication-only, and deletion-only models, respectively. These were combined into 70 448 risk-increasing (i.e., no disease-protecting CNV) associations and 3 disease burden 449 associations (Figure 2; Table S3). Forty-five associations (45/73 = 62%) were supported at 450 GW significance by multiple models, the lowest p -value (i.e., “best model”) being obtained 451 through the mirror, deletion-only, U-shape, and duplication-only models for 24, 23, 21, and 5 452 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 18 of the signals, respectively. No association was detected at GW significance by both the 453 duplication-only and deletion-only models, so that each signal was attributed a “main model” 454 that indicates whether the association is primarily driven by duplications or deletion s (see 455 Methods; Figure 2). The main model should be interpreted with caution as both deletions and 456 duplications might influence disease risk but only one CNV type -specific model might reach 457 GW significance (e.g., due to higher frequency) . This is particularly relevant as 73% (33/45) 458 of disease-associated CNV regions ( CNVRs) have a hig her duplication than deletion 459 frequency (Figure 2A). Hence, 95% (20/21) of signals mainly driven by duplications were also 460 identified by the mirror/U-shape model(s) and contribution of deletions cannot be excluded. 461 Figure 2. CNV-disease association map 462 (A) Duplication and deletion frequencies ([%]; y-axis; break: //) of the lead probe for each unique and 463 non-overlapping disease-associated CNV region (CNVR), labeled with corresponding cytogenic band 464 (x-axis; 16p11.2 is split to distinguish the distal 220kb breakpoint BP2 -3 and proximal 600kb BP4 -5 465 CNVRs; non-overlapping CNVRs on the same cytogenic band are numbered). If signals mapping to 466 the same CNVR have different lead probes, the maximal frequency was plotted. ( B) Associations 467 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 19 between CNVRs (x-axis) and diseases (y-axis) identified through CNV-GWAS. Color indicates the main 468 association model. Size and transparency reflect the statistical confidence tier. Black contours indicate 469 overlap with OMIM gene causing a disease with shared phenotypic features. Black crosses indicate 470 overlap with SNP-GWAS signal for a related trait. Grey shaded lines indicate CNVRs with continuous 471 trait associations [23]. N provides count for various features. 472 Validation of identified CNV-GWAS signals 473 Across the 45 CNVRs, CNV frequencies were low, ranging between 0.01% ( our frequency 474 cutoff) and 0.36%, with 87% ( 39/45) of CNVRs having a frequency ≤ 0.1% ( Figure 2 A). 475 Consequently, associations rely on a low number of diseased CNV carriers and require 476 validation (see Methods; Figure 1A; bottom; Figure 2B; Table S3). We used three statistical 477 approaches to assess the robustness of CNV-diseases associations: i) Fisher test, ii) residual 478 regression, and iii) time -to-event analysis through CoxPH modeling. We replicated 28/70 479 (40%), 23/70 (33%), and 70/70 (100%) of the associations with the respective methods at the 480 arbitrary validation threshold of p ≤ 10 -4. This allowed to stratify associations in confidence 481 tiers, with 17 signals replicating with all methods (tier 1), 20 with two (tier 2), and 36 only 482 through time-to-event analysis (tier 3). Importantly, time-to-event analysis showed that CNVs 483 always contributed to an earlier age of disease onset, in line with the paradigm that diseases 484 with a strong genetic etiology have earlier onset [53]. 485 486 In parallel, we gathered literature evidence linking genetic variation at CNVRs with relevant 487 phenotypes (Table S3). Forty-eight signals (48/73 = 64%) mapped to a CNVR harboring a 488 least one OMIM morbid gene and in 15 cases, the gene was linked to a Mendelian disorder 489 sharing phenotypic features with the associated common disease . For instance, association 490 between 4q35 CNVs and corneal conditions (chr4:186,687,554-187,182,384; ORU-shape = 18.2; 491 95%-CI [5.2; 63.1]; p = 5.0 x 10-6) encompasses CYP4V2 [MIM: 608614], a gene associated 492 with autosomal recessive Bietti crystalline corneoretinal dystrophy [MIM: 210370], a disorder 493 that impairs vision and progresses to blindness by age 50-60 years [54]. We next assessed 494 whether SNPs overlapping disease-associated CNVRs were reported to associate with the 495 implicated disease or a biomarker thereof in the GWAS Catalog. This was the case for 28 496 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 20 (28/66 = 42%) autosomal signals, a similar proportion (38%) than for continuous trait CNV -497 GWAS [23]. For instance, distal 22q11.2 CNVs increased risk for disorders of mineral 498 metabolism (chr22:21,797,101-22,661,627; ORmirror = 0.02; 95%-CI [0.006; 0.083]; p = 9.9 x 499 10-9) and overlap ped heel bone mineral density SNP -GWASs signals, while 3q29 CNVs 500 increased Alzheimer’s disease risk (chr3:196,953,177-197,331,898; ORU-shape = 11.8; 95%-CI 501 [4.0; 34.7]; p = 6.6 x 10 -6) and overlap ped with SNP-GWAS signal for PHF -tau levels, and 502 suggestive sig nals (p < 5 x 10 -6) for frontotemporal dementi a and cognitive decline in 503 Alzheimer’s disease. Finally, 37 signals (37/73 = 51%) mapped to nine CNVRs previously 504 found to be associated with complex traits [23]. 505 506 We also set out to replicate association signals in 90,211 unrelated EstBB individuals [40], 507 using similarly case definition than for the UKBB (see Methods; Figure S1B). Requesting at 508 least one diseased CNV carrier, 33 of 73 associations could be evaluated, among which four 509 were strictly replicated (p ≤ 0.05/33 = 1.5 x 10 -3) and five additional ones reached nominal 510 significance (p ≤ 0.05) (Table S3). Compared to what would be expected by chance, t his 511 corresponds to a 5.5-fold (pbinomial = 2.5 x 10-5) and 30.3-fold (pbinomial = 6.6 x 10-7) enrichment 512 for replication at p ≤ 0.05 and p ≤ 5 x 10-3, respectively (Figure 3A). Despite low power, these 513

support validity of the primary UKBB association signals. Signals replicating at nominal 514 significance are detailed in Figure 3B . Many harbor SNP -GWAS signals for related 515 phenotypes (7/9), relevant morbid OMIM genes (4/9), or map to CNVRs previously associated 516 with similar diseases (6/ 7) or biomarkers (4/9). Among them, two are in the lowest UKBB 517 confidence tier. 15q13 duplications were linked to increased risk for acute kidney injury (AKI; 518 chr15:30,946,160-31,881,106 | UKBB: ORdup = 4.6; 95%-CI [2.5; 8.4]; p = 7.1 x 10-7 | EstBB: 519 p = 2.4 x 10-4). Homozygous mutations in FAN1 [MIM: 613534], one of the five genes mapping 520 to this CNVR, have been linked to karyomegalic interstitial nephritis [MIM: 614817], a 521 progressive renal condition that leads to CKD [55]. The second example links CNVs affecting 522 exon 2 and intron 2-3 of PRKN ([MIM: 602544]) – a gene causing juvenile autosomal recessive 523 Parkinson’s disease [MIM: 600116] – to sleep disorders such as insomnia and hypersomnia 524 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 21 (chr6:162,705,164-162,873,489 | UKBB: ORmirror = 0.12; 95%-CI [0.05; 0.26]; p = 1.6 x 10 -7 | 525 EstBB: p = 0.047). This finding is particularly relevant given the region’s high CNV frequency 526 (0.22%) and the fact that sleep disturbances are among the earliest symptoms of Parkinson’s 527 disease [56]. Follow-up studies should determine whether these individuals are more prone 528 to develop Parkinson’s disease in the future. 529 Figure 3. Replication of CNV-disease associations in the Estonian Biobank 530 (A) Enrichment for signal replication (y-axis; 95% confidence interval as grey ribbon) at different levels 531 of significance (alpha; x-axis) in the Estonian Biobank (EstBB). Color and size indicate the p -value of 532 the enrichment (one -sided binomial test) and the number of observed associations, respectively. 533 Dashed red line indicates 1x enrichment, i.e., the number of observed associations matches the number 534 of expected ones. ( B) Associations replicated at nominal significance in the EstBB , color -stratified 535 according to whether they meet the discovery genome-wide (GW; p ≤ 7.5 x 10-6; dark green), replication 536 (p ≤ 1.5 x 10-3; green), or nominal (p ≤ 0.05; light green) significance threshold. Disease (CKD = chronic 537 kidney disease; AKI = acute kidney injury; HTN = hypertension; COPD = chronic obstructive lung 538 disease), cytogenic band and coordinates, best model (MIR = mirror; U = U-shape; DUP = duplication-539 only; DEL = deletion-only), odds ratio (OR), p-value (P) and statistical confidence tier are given for the 540 UK Biobank (UKBB) discovery analysis. OR, one -sided p-values, and number of cases among CNV 541 carriers are provided for the EstBB replication. Overlap with SNP-GWAS signals for a related trait (✓ = 542 yes; ✗ = no) or a relevant OMIM gene (RCAD = renal cyst and diabetes; KIN = karyomegalic interstitial 543 nephritis; PXE = pseudoxanthoma elasticum; PD = Parkinson’s disease ) is indicated. Previous 544 association with diseases[20] (duplication (DUP) or deletion (DEL) was associated with indicated 545 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 22 disease; no association (✗); some CNVRs were not tested) and continuous traits [23] (disease-relevant 546 biomarkers are specified; other traits (*); no association (✗)) are listed. 547 Evidence provided by statistical, literature-based, or independent replication help prioritizing 548 the most promising associations for follow-up studies and pinpoint plausible candidate genes. 549 We highlight several examples where deviations by one copy-number are linked to common 550 diseases sharing clinical features with rare Mendelian conditions caused by homozygous 551 perturbations of the same genetic region. This argues against a dichotomic view on dominant 552 versus recessive modes of inheritanc e and analogously to allelic series, suggest that 553 Mendelian and common diseases represent different ends of the phenotypic spectrum caused 554 by genetic variation at a given locus. 555 556 Global characterization of disease-associated CNV regions 557 We sought to identify global characteristics that distinguish disease-associated CNVRs (Table 558 S4). Number of p rotein-coding genes embedded in disease -associated CNVRs, hereafter 559 referred to as “disease genes”, ranged from 0 to over 30 and generally correlated with the 560 number of encompassed probes ( rPearson = 0.50; p = 4.2 x 10 -4; Figure S3A). Exceptions 561 include single-gene CNVRs overlapping known pathogenic genes captured thanks to high 562 probe coverage (e.g., BRCA1). While only seven CNVRs (16%) associated with multiple 563 diseases, propensity for pleiotropy depended on CNV length (+0.16 association/disease gene; 564 p = 1.5 x 10-5). Accordingly, CNVRs containing more than five genes were also more likely to 565 associate with continuous traits (ORFisher = 53.2; p = 8.5 x 10-6) [23]. One CNVR that stood out 566 is the 600kb 16p11.2 BP4-5 region (Figure 2B). Originally identified as a major risk factor for 567 autism, schizophrenia, developmental delay and intellectual disability , macro-/microcephaly, 568 epilepsy, and obesity [57–63], we previously found the region to associate with 26 continuous 569 complex traits [23]. Here, we show that 16p11.2 BP4 -5 deletions increase the risk of 12 570 diseases – including both new and previously reported associations across multiple organ 571 systems – as well as the disease burden (+3 diseases/deletion; p = 1.2 x 10 -26), while the 572 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 23 region’s duplication drove increased risk for psychiatric conditions (i.e., bipolar disorder, 573 schizophrenia, and depression), in line with previous findings [62]. 574 575 Next, we assessed whether disease genes were under stronger evolutionary constraint (i.e., 576 less tolerant to mutations) than genes affected by CNVs at the same frequency but not 577 associated with any disease (i.e., “background genes”). Compared to background genes, the 578 231 disease genes had more constrained pLI (pWilcoxon = 1.3 x 10-4; Figure S3B) and LOEUF 579 (pWilcoxon = 1.9 x 10-7; Figure S3C) scores, suggesting stronger intolerance to LoF mutations. 580 Splitting CNVRs depending on whether they have at least one association through either the 581 duplication-only or deletion-only model, we evaluated whether embedded disease genes were 582 sensitive to having less (i.e, haploinsufficiency; Figure S3D) or more (i.e., triplosensitivity; 583 Figure S3E) than two functional copies. No significant difference in pHaplo score s were 584 observed but genes overlapping regions whose duplication (pWilcoxon = 9.0 x 10-19) and deletion 585 (pWilcoxon = 1.0 x 10-23) have been linked to diseases were more likely to be triplosensitive than 586

genes . Similar tr ends were observed considering genes overlapping CNVRs 587 involved uniquely through the duplication -only and deletion -only models and not the other 588 CNV type -specific model (Figure S 3F-G). Overall, our results indicate that a CNVR’s 589 pathogenicity is determined both by the number and characteristics of affected genes. 590 591 New insights in known disease genes 592 Two out of 12 female BRCA1 deletion carriers were diagnosed with ovarian cancer 593 (chr17:41,197,733-41,276,111; ORdel = 284.3; 95%-CI [24.6; 3290.8]; p = 6.1 x 10 -6; Figure 594 4A). BRCA1 [MIM: 113705] is a tumor suppressor gene whose LoF represents a major genetic 595 risk factor for the development of hereditary breast and ovarian cancer (HBOC) [MIM: 604370] 596 [64]. Exploring the clinical records of the 12 deletion carriers, we found five diagnoses of breast 597 cancer (a trait assessed by CNV-GWAS but that did not yield a GW -significant association), 598 one of endometrial cancer, and one of Fallopian tube cancer, so that eight carriers (67%) had 599 received a HBOC diagnosis (Figure 4B). Not only was prevalence of HBOC higher among 600 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 24 BRCA1 deletion carriers (ORFisher = 31.0; p = 1.1 x 10-6), but disease onset was earlier (HR = 601 17.0; p = 1.3 x 10 -15; Figure 4C). Among the four carriers with no HBOC, two had received 602 cancer prophylactic surgery, de facto reducing the penetrance of the deletion. Surgeries were 603 likely carried out based on family history of HBOC, which was reported for 6 carriers (50%), 604 suggesting that these deletions are inherited. We did not observe higher prevalence of other 605 cancer types (Figure 4B). 606 607 High abundance of Alu repeats make the low-density lipoprotein (LDL) receptor (LDLR) [MIM: 608 606945] susceptible to CNVs [65]. We found that deletion of exon 2 -6 increased risk for 609 ischemic heart disease (IHD; chr19:11,210,904-11,218,188; ORdel = 31.2; 95%-CI [7.1; 137.8]; 610 p = 5.6 x 10-6), a condition present in 8 of 14 deletion carriers (Figure 4D). Heterozygous - and 611 less frequently homozygous – mutations in LDLR represent the main genetic etiology for 612 familial hypercholesterolemia [66], which is characterized by elevated LDL cholesterol and 613 predisposition for adverse cardiovascular outcomes [67]. Previously identified in clinical 614 studies of familial hypercholesterolemia [68], the CNVR implicated by our analysis specifically 615 encompasses the ligand-binding domain of LDLR [66]. Confirming widespread prevalence and 616 family history (43%) of cardiovascular diseases (Figure 4E), medical records of deletion 617 carriers further revealed higher prevalence (ORFisher = 11.6; p = 7.9 x 10-5) and earlier onset 618 (HR = 5.8; p = 1.4 x 10-7; Figure 4F) of pure hypercholesterolemia (E78.0), a code included in 619 our lipidemia definition but that did not yield a signal pick -up by the CNV -GWAS. As we 620 previously did not find the CNVR to associate with standardized blood biochemistry LDL levels 621 [23], we hypothesized that the latter were lowered by hypolipidemic agents. Ten (71%) 622 deletion carriers were on statins and six (43%) were additionally using cholesterol absorption 623 inhibitors, while the remaining four did not receive a dyslipidemia or IHD diagnosis and 624 harbored smaller deletions (i.e., P12-14; Figure 4E). We concluded that drugs likely masked 625 genetically determined LDL levels, as shown by higher LDL levels in the first primary care 626 measurement on record, measured prior to the standardized LDL measurement (pt-test = 0.03; 627 Figure 4G). Despite this, the recommended target of ≤ 1.8 mmol/L for high-risk individuals [69] 628 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 25 was never met. By recovering known gene-disease pairs typically studied in clinical cohorts, 629 we showcase how the rich phenotypic data from biobanks can generate insights into the 630 mechanisms, epidemiology, and comorbidities of these diseases, implicating CNVs as 631 important genetic risk factors. 632 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 26 Figure 4. Refining contribution of CNVs to gene-disease pairs 633 (A) Genomic coordinates of the 12 females (P1 -12) carrying a BRCA1 deletion (CNVR delimited by 634 vertical dashed lines ), colored according to ovarian cancer diagnosis. ( B) Left: Cancer and related 635 family/personal diagnoses received by individuals in (A). Color indicates age at diagnosis. Right: Counts 636 per ICD-10 code. (C) Kaplan-Meier curve depicting the percentage, with 95% confidence interval, of 637 females free of female-specific cancers over time among copy -neutral and BRCA1 deletion carriers. 638 Hazard ratio (HR ) and p -value for the BRCA1 deletion are given (CoxPH model). ( D) Genomic 639 coordinates of the 14 individuals (P1-14) carrying an LDLR deletion (CNVR delimited by vertical dashed 640 lines), colored according to ischemic heart disease (IHD) diagnosis. ( E) Left: Medical conditions and 641 family/personal diagnoses and medication received by ≥ 3 LDLR deletion carriers in (D) , following 642 legend in (B). (F) Kaplan-Meier curve for pure hypercholesterolemia (E78.0) among copy-neutral and 643 LDLR deletion carriers, following legend as in (C). (G) Low density lipoprotein (LDL)-cholesterol levels 644 (y-axis) from primary care data (first available measurement) and blood biochemistry (average over 645 instances) for six deletion carriers in (D) with at leas t one antecedent primary care LDL -cholesterol 646 measurement, colored according to IHD diagnosis. P -value compares the two data sources (paired 647 one-sided t -test). Grey horizontal line represents median LDL -cholesterol value (from blood 648 biochemistry) in non -carriers. Light and darker green background represent recommended target 649 values for low (≤ 3 mmol/L) and high (≤ 1.8 mmol/L) risk individuals, respectively. (H) 17q12 association 650 landscape. Top: Negative logarithm of association p -values of CNVs (dark grey; CNVR delimited by 651 vertical dashed lines) and SNPs (orange) [70] with chronic kidney disease (CKD) and SNPs with 652 estimated glomerular filtration rate (eGFR; red) [71]. Lead SNPs are labeled. Red horizontal dashed 653 lines represent the genome-wide threshold for significance for CNV-GWAS (p ≤ 7.5 x 10-6) and SNP-654 GWAS (p ≤ 5 x 10 -8). Middle: Genomic coordinates of genes and DECIPHER CNV, with HNF1B, the 655 putative causal gene in red. Segmental duplications are represented as a gray gradient proportional to 656 the degree of similarity. Bottom: Genomic coordinates of duplications (blue) and deletions (red) of UK 657 Biobank participants overlapping the region. (I) CKD prevalence (± standard error) according to 17q12 658 copy-number (CN). P -values compare deletion (CN = 1) and duplication (CN = 3) carriers to copy -659 neutral (CN = 2) individuals (two-sided Fisher test). Number of cases and samples sizes are indicated 660 (N = cases/sample size). (J) eGFR levels according to 17q12 CN, shown as boxplots; outliers are not 661 shown. P-values comparisons as in (I) (two-sided t-test). Grey horizontal line represents median eGFR 662 in non-carriers. Light and darker green background represent mildly decreased (60-90 ml/min/1.73m2) 663 and normal (≥ 90 ml/min/1.73m2) kidney function, respectively. (K) Kaplan-Meier curve for CKD among 664 17q12 deletion and duplication carriers, following legend as in (C). 665 Biomarker CNV associations tag pathophysiological processes 666 Integration of biomarker and disease CNV -GWAS signals can identify high -confidence, 667 clinically relevant associations. Heterozygous LoF of HNF1B [MIM: 189907] and 17q12 668 deletions cause renal cyst and diabetes (RCAD) [MIM: 137920 ], a severe disorder 669 characterized by renal abnormalities and maturity-onset diabetes by the young [72,73]. While 670 we previously showed that renal biomarkers were increased in duplication carriers [23], here, 671 we demonstrate that both 17q12 deletions and duplications increase CKD risk 672 (chr17:34,755,219-36,249,489; ORU-shape = 6.5; 95%-CI [3.4; 12.1]; p = 5.9 x 10-9; Figure 4H), 673 with a prevalence of 33.3% (pt-test = 0.026) and 16.9% (pt-test = 6.8 x 10-5) among deletion and 674 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 27 duplication carriers respectively, versus 4.4% in copy-neutral individuals (Figure 4I). Results 675 replicated in the EstBB (p = 1.8. x 10-7; Figure 3B) and are supported by 20% of CNV carriers 676 showing signs of kidney disease based on estimated glomerular filtration rate (eGFR < 6 0 677 ml/min/1.73m2), compared to 2.2% in copy -neutral individuals (Figure 4J). Importantly, both 678 17q12 deletion and duplication lower age of CKD onset (HR ≥ 4.6; p ≥ 1.3 x 10 -7; Figure 4K), 679 providing strong evidence of the deleterious consequences on kidney health of altered dosage 680 of 17q12. 681 682 Similarly, the blood pressure -increasing 16p12.2 deletion (chr16: 21,946,523-22,440,319) 683 [19,23] increased risk for hypertension (OR del = 2.7; 95% -CI [1.9; 3.8]; p = 1.3 x 10 -8) and 684 cardiac conduction disorders (ORdel = 3.3; 95%-CI [2.2; 4.9]; p = 1.1 x 10-8), suggesting a role 685 in cardiovascular health (Figure S4A-D). Primarily associated with developmental delay and 686 intellectual disability [74,75] – proxied by decreased fluid intelligence (pt-test = 8.7 x 10 -5) and 687 income (pt-test = 1.4 x 10-12) in the UKBB (Figure S4E-F) – cardiac malformations are reported 688 in ~38% of clinically ascertained cases [76]. Among 193 UKBB deletion carriers, two (1%) had 689 congenital insufficiency of the aortic valve (Q23.1), correspo nding to a higher but not 690 significantly different prevalence of cardiovascular malformations (Q20 -28) than in copy -691 neutral individuals (ORFisher = 2.1; p = 0.251). The deletion also associated with pneumonia 692 (ORdel = 3.0; 95%-CI [1.9; 4.6]; p = 5.4 x 10 -7), coinciding with associations with decreased 693 forced vital capacity [23] (Figure S4G-H) and peak expiratory flow [19], together demonstrating 694 the clinical relevance of CNV-biomarker associations. 695 696 Dissecting complex pleiotropic CNV regions 697 While some CNV signals converge onto the same underlying physiological processes, others 698 tie apparently unrelated traits to the same genetic region, suggesting genuine pleiotropy. 699 16p13.11 harbors multiple, partially overlapping recurrent groups of CNVs allowing fine-700 mapping of signals to different subregions of the CNVR (Figure 5). Through different 701 association models, the CNVR was linked to uncorrelated traits including epilepsy, kidney 702 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 28 stones, hypertension, alkaline phosphatase (ALP), forced vital cap acity, and age at 703 menopause and menarche . We previously proposed MARF1 as a candidate gene for the 704 female reproductive phenotypes [23] and will focus here on the remaining traits. 705 Figure 5. Dissection of complex pleiotropic patterns of recurrent CNVs at 16p13.11 706 (A) 16p13.11 genetic landscape. Coordinates of UK Biobank duplications (shades of blue; top) and 707 deletions (shades of red; bottom) overlapping the maximal CNV region (CNVR delimited by vertical 708 dashed lines) associated with epilepsy, kidney stones, hypertension, and alkaline phosphatase (ALP). 709 CNVs are divided and colored according to five categories (cat1-5) to reflect recurrent breakpoints, with 710 atypical CNVs in grey. Breakpoints reflect segmental duplications, represented with a grey gradient 711 proportional to the degree of similarity. Middle: genomic coordinate s of genes and DECIPHER CNV. 712 Inset: Overlap between ABCC6’s exonic structure and cat5 deletions. (B, D, F, H) Negative logarithm 713 of association p-values of CNVs with ( B) epilepsy; (D) kidney stones; ( F) hypertension; and (H) ALP 714 (dark grey; model in parenthesis; CNVR delimited by vertical dashed lines) and SNPs with (B) epilepsy 715 [77]; ( D) kidney stones [78], calcium levels, and phosphate levels (y -axis; break: //); ( F) essential 716 hypertension and systolic blood pressure [79]; and (H) ALP. Lead SNPs are labeled. Red horizontal 717 dashed lines represent genome-wide thresholds for significance for CNV -GWAS (p ≤ 7.5 x 10 -6) and 718 SNP-GWAS (p ≤ 5 x 10-8). (C, E, G) Prevalence (± standard error) of (C) epilepsy, (E) kidney stones, 719 and (G) hypertension according to 16p13.11 copy-number (CN) and CNV categories from (A). P-values 720 compare deletion (CN = 1) and duplication (CN = 3) carriers to copy-neutral (CN = 2) individuals (two-721 sided Fisher test). Number of cases and samples sizes are indicated (N = cases/sample size). (I) ALP 722 levels according to 16p13.11 CN and CNV category, shown as boxplots; outliers are not shown. P -723 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 29 values compare deletion (CN = 1) and duplication (CN = 3) carriers to copy-neutral (CN = 2) individuals 724 (two-sided t-test). Grey horizontal line represents median ALP value in non-carriers. 725 The 654 duplications and 355 deletions overlapping the maximal CNVR (chr16 :15,070,916-726 16,353,166) were grouped into 5 categories (cat1-5) based on their breakpoints (Figure 5A; 727 Supplemental Note 3). Risk for epilepsy was increased in deletion carriers (chr16:15,122,801-728 16,353,166; ORdel = 6.2; 95%-CI [2.8; 13.4]; p = 4.4 x 10 -6; Figure 5B), with a prevalence of 729 8.3% among cat1-4 deletion carriers compared to less than 1.5% among copy -neutral and 730 duplication carriers ( Figure 5C). Previously associated with epilepsy in clinical cohorts 731 [6,80,81], t he region harbors NDE1 [MIM: 609449], a ge ne associated with autosomal 732 recessive lissencephaly [MIM: 614019] and microhydranencephaly [MIM: 605013] and whose 733 mutation has been linked to epilepsy [82,83]. Deletions also increased risk for kidney stones 734 (chr16:15,120,501-16,353,166; ORdel = 5.9; 95%-CI [2.9; 11.9]; p = 7.3 x 10-7), with the CNV-735 GWAS signal peaking close to a missense variant (rs41278174 G>A; FrequencyA: 2.1%) in 736 exon 23 of ABCC6 [MIM: 603234] associating with calcium and phosphate levels through 737 SNP-GWAS (Figure 5D). These signals coincide with the recurrent cat5 deletion that covers 738 29 probes spanning exons 23-29 from ABCC6 (Figure 5A). Kidney stones prevalence reaches 739 4.3% among cat5 deletion carriers, in-between estimates for larger cat1 -4 deletion carriers 740 (9.3%) and copy -neutral individuals (2.3%) (Figure 5E). A wide range of variants affecting 741 ABCC6 have been identified and linked to the calcification disorder pseudoxanthoma 742 elasticum through recessive [MIM: 264800] – and more rarely dominant [MIM: 177850] – 743 inheritance [84–87], with the Alu-mediated cat5 deletion representing one of the most frequent 744 variants [88,89]. ABCC6 is expressed in the kidney and recent estimates from clinical cohorts 745 suggested that kidney stone s are an unrecognized (i.e., not used to establish clinical 746 diagnosis) but prevalent (11-40%) feature of pseudoxanthoma elasticum [90–92]. Our data 747 support kidney stones as a clinical outcome of ABCC6 disruption with partial gene deletions 748 leading to lower penetrance than large 16p13.11 deletions. Unlike epilepsy and kidney stones, 749 both deletion (39.2%) and duplication (42%) carriers are at increased risk for hypertension 750 (chr16:15,127,986-16,308,285; ORU-shape = 1.5; 95%-CI [1.3; 1.8]; p = 5.5 x 10-6; Figure 5F), 751 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 30 compared to copy-neutral individuals (35.3%) (Figure 5G). The CNVR overlaps a SNP-GWAS 752 signal for systolic blood pressure mapping to MYH11 [MIM: 160745] (Figure 5F). Expressed 753 in arteries, MYH11 encodes for smooth muscle myosin heavy chains and has been linked to 754 dominant familial thoracic aortic aneurysm [MIM: 132900], for which hypertension represents 755 a leading risk factor. Increased prevalence (37.4%) of hypertension among cat5 deletions 756 implicates ABCC6, suggesting cis-epistasis for hypertension risk at 16p13.11. Consistent with 757 this model, ABCC6 plays a role in vascular calcification as the causal gene for generalized 758 arterial calcification of infancy [MIM: 614473] [93,94], typically diagnosed by hypertension in 759 newborns. Interestingly, the previously described mirror association with ALP 760 (chr16:15,070,916-16,276,964; bmirror = 6.6 U/L; p = 3.5 x 10 -7) peaks at the distal end of the 761 CNVR [23], nearby a suggestive SNP-GWAS signal for ALP levels (Figure 5H). Splitting ALP 762 levels by CNV category revealed that th is mirroring effect is driven by individuals with cat2 763 deletion (mean = 76.4 U/L; pt-test = 9.7 x 10-3) and duplication (mean = 92.9 U/L; pt-test = 8.2 x 764 10-5), as other CNV carriers had ALP levels indistinguishable from those of copy -neutral 765 individuals (mean = 83.6 U/L) (Figure 5I). Hence, we propose the distal region of the CNVR 766 to harbor the critical region regulating ALP levels, even if no obvious candidate gene could be 767 identified through literature review. 768 769 The proximal 22q11.2 region , previously linked to DiGeorge [MIM: 188400] and 770 velocardiofacial [MIM: 192430] syndromes, harbors four low-copy repeat (LCR; labeled A-to-771 D) [95]. Building on evidence of complex association patterns with this CNVR [38], we report 772 novel associations between CNVs spanning LCR A -D and IHD (chr22: 19,024,651-773 21,463,545; OR U-shape = 2.1; 95% -CI [1.6; 2.8]; p = 1.5 x 10 -7), LCR B -D and aneurysm 774 (chr22:20,708,685-21,460,008; ORdel = 41.8; 95%-CI [10.0; 175.1]; p = 3.2 x 10 -7), and LCR 775 A-C and headaches (chr22:19,024,651-21,110,240; ORmirror. = 3.7; 95%-CI [2.1; 6.5]; p = 4.8 776 x 10 -6) (Figure S 5A; Supplemental Note 3). Based on 3 LCR B -D deletion carriers with 777 aneurysm, this corresponds to a 22-times higher prevalence than in copy-neutral individuals 778 (Figure S5B). Association with IHD is better powered, with a prevalence of 12%, 21%, 16%, 779 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 31 and 20% among copy -neutral individuals and carriers of LCR C -D, B -D, and A -D CNVs, 780 respectively (Figure S5C). This suggests that IHD risk scales with the amount of affected 781 genetic content, supporting the presence of genetic driver (s) and/or modifier(s) in the C -D 782 interval, beyond the prime candidate TBX1 [MIM: 602054] [95]. Collectively, our data indicates 783 that altered 22q11.2 dosage can result in a spectrum of cardiovascular afflictions of various 784 degrees of severity, ranging from well-described congenital malformation [95,96] to adult -785 onset cardiovascular disorders. 786 787 15q13 deletions spanning BP4-5 [MIM: 612001] – and to a lesser extent duplications – have 788 been associated with neuropsychiatric and developmental conditions [97,98], with the nicotinic 789 acetylcholine receptor ion channel CHRNA7 being proposed as the driver gene based on the 790 presence of similar phenotypes in individuals with a smaller deletion (D-CHRNA7-BP5) only 791 affecting CHRNA7 [99] (Figure S 6A). While BP4 -5 duplication carriers showed higher 792 prevalence of AKI (EstBB-replicated: Figures 3B, S6B), hemorrhagic stroke 793 (chr15:30,912,719-31,982,408; ORU-shape = 7.5; 95%-CI [3.2; 17.9]; p = 4.3 x 10-6; Figure S6C), 794 and anemia (chr15:30,912,719-31,094,479; ORdup = 4.9; 95% -CI [2.5; 9.7]; p = 3.2 x 10 -6; 795 Figure S6D), reminiscent of associations with pulse rate, mean corpuscular hemoglobin, and 796 red blood cell count [19,23], this was not the case for the ~10 -times more numerous D-797 CHRNA7-BP5 duplication carriers. Replicating an association with a sthma [20] 798 (chr15:30,912,719-32,516,949; OR mirror = 0.17; 95% -CI [0.08; 0.35]; p = 1.2 x 10 -6) which 799 parallels decreased forced vital capacity [23] and peak expiratory flow [19], this was the only 800 deletion-driven signal for which the CNVR encompassed the entire BP4-5 region. However, 801 only BP4-5 (46.2%; p t-test = 1.8 x 10 -5) deletion carriers had higher asthma prevalence than 802 copy-neutral individuals (12.1%) ( Figure S 6E). Overall, the non -neurological disorders we 803 associate with 15q13 CNVs appear to specifically involve dosage of the genes within BP4-D-804 CHRNA7 and not CHRNA7. 805 806 Pathological consequences of an increased CNV burden 807 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 32 By assessing the global pathogenic impact of CNVs we can capture the effect of ultra-rare 808 variants (frequency ≤ 0.01%), as well as those whose effect is not strong enough to reach GW 809 significance under current settings. Individual-level autosomal CNV (duplication + deletion), 810 duplication, and deletion burdens were calculated as the number of Mb or genes affected by 811 the considered type of variation and their predictive value on the same 60 diseases (and the 812 disease burden) previously assessed through CNV -GWAS was estimated ( Figure 6A; left). 813 Disease burden strongly associated with a high CNV load (Mb: bCNV = +0.08 disease/Mb; p = 814 3.7 x 10-27 | Gene: bdel = +0.03 disease/gene; p = 3.7 x 10-27) and risk for 20 individual disorders 815 was increased by at least one type of CNV burden (p ≤ 0.05/61 = 8.2 x 10-4; Figure 6B; left; 816 Table S5). Strongest effect sizes were observed for the Mb burden for psychiatric disorders, 817 such as bipolar disorder (ORdel = 1.4; p = 6.9 x 10-4), schizophrenia (ORdel = 1.4; p = 4.1 x 10-818 5), or epilepsy (ORCNV = 1.1; p = 8.3 x 10 -5), in agreement with CNVs representing important 819 risk factors for these complex and polygenic disorders. To ensure that we do not merely 820 capture the effect of individual CNV-disease associations previously isolated by CNV-GWAS, 821 we excluded CNVs overlapping disease-associated CNVRs from the burden calculation and 822 re-estimated their predictive value (Figure 6A; right). Overall association strength dropped but 823 signal was lost only for type 1 diabetes and chronic obstructive pulmonary disease (Figure 6B; 824 right; Table S 5), s uggesting that CNVs not captured by the CNV-GWAS contribute to 825 modulating disease risk. Establishing the polygenic CNV architecture of a substantial number 826 of common diseases , this implies that increased power will likely lead to the discovery of 827 further CNV-disease associations. 828 829 As the “gene burden” captures more associations than the “Mb burden” and as the deletion 830 burden yields stronger associations than the duplication one (Figure 6B), we hypothesized 831 that the pathogenic effect of CNVs is mainly driven by the number of deleted genes. Based 832 on an average GW gene density of 8.4 genes/Mb, we assume the deletion burden measured 833 in Mb (𝛽:;) to be 8.4 times larger than the one measured in number of affected genes (𝛽<"!"), 834 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 33 expecting a ratio of 8.4 if the type of affected genetic content does not impact disease risk 835 (see Methods; Figure 6C). Ratios significantly (p ≤ 0.05/21 = 2.4 x 10-3) smaller or larger than 836 8.4 indicate disproportionately large or small contribution of protein coding regions, 837 respectively. We detected a larger than expected contribution of deleted genes to the disease 838 burden (ratio = 4.4; p = 1.3 x 10-6), and AKI (ratio = 3.5; p = 7.0. x 10-4) and lipidemia (ratio = 839 2.7; p = 1.4 x 10 -3) risk, along with seven additional nominally decreased ratios ( Figure 6D; 840 left; Table S6), confirming that haploinsufficient genes mediate the burden’s effect on disease 841 risk. Repeating the analysis with the duplication burden did not reveal any significant 842 deviations ( Table S 6). After correcting for CNV -GWAS signals, only AKI (p = 7.0 x 10 -4), 843 arthrosis (p = 0.030), and disease burden (p = 0.016) showed increased contribution of deleted 844 genes on disease risk, indicating that a substantial fraction of effects capture by the CNV -845 GWAS are mediated by deletion of coding regions. 846 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 34 Figure 6. The coding deletion burden increases overall disease risk 847 (A) Burden calculation. Left: Raw CNV burden is calculated by summing up the length (in Mb or number 848 of affected genes) of all CNVs in an individual. Burden values are used as a predictor for disease risk. 849 Right: Corrected CNV burden is calculated by summing up the length of all CNVs that do not overlap 850 with a CNV region (CNVR) significantly associated with the investigated disease through CNV-GWAS. 851 Corrected burden values are used to re -estimate contribution of the CNV burden to disease risk (red 852 curve). (B) Contribution of raw and corrected CNV burdens in Mb or affected genes (x-axis) to disease 853 risk (y-axis). Only the most significantly associated burden, providing p ≤ 0.05/61 = 8.2 x 10-4, is shown. 854 Color indicates whether the CNV (duplication + deletion), duplication, or deletion burden was most 855 significantly associated, with size and transparency being proportional to the effect size (beta) and p -856 value, respectively. Grey horizontal bands mark t raits with no CNV-GWAS signal. (C) Top: Based on 857 an average genome-wide gene density of 8.4 genes/Mb, we expect the effect of the CNV burden in Mb 858 to be 8.4x larger than the one measured in number of affected genes. Bottom: If the ratio is smaller 859 than 8.4, it indicates a n increased contribution of coding regions. Alternatively, if larger than 8.4, it 860 indicates a decreased contribution of coding regions. ( D) Ratio, with 95% confidence interval (CI), of 861 the raw and corrected Mb/genes deletion burdens (x -axis) for all diseases showing at least one 862 significant burden association. Significant deviation from the expected ratio of 8.4 (grey vertical line) is 863 indicated as *** for p ≤ 0.05/21 = 2.4e-3 and * for p ≤ 0.05 (two-sided one sample t-test). 864 865

866 Using an adapted GWAS framework, we provide a detailed investigation of the contribution of 867 CNVs to the genetic architecture of 60 common diseases and showcase how the rich 868 phenotypic data of the UKBB can be leveraged to gain new biological insights, highlighting the 869 role of CNVs as modulators of common disease susceptibility in the general population. 870 871 Various strategies have been used to study CNV-disease associations in the UKBB. Focusing 872 on diseases related to the ones assessed in the current study, we replicate 10 out of the 24 873 detected associations (at FDR ≤ 0.1) with 54 likely pathogenic CNVs [20] and all four 874 associations (at p ≤ 1 x 10-9) in a recent CNV-GWAS investigating 757 diseases [25]. Despite 875 analyzing the same dataset, we often obtained p-values orders of magnitude smaller (e.g., 876 16p11.2 BP4-5 deletion and AKI: p = 5.6 x 10 -20; pCrawford = 6.0 x 10 -5; pHujoel = 3.3 x 10 -15). 877 Besides accruing case count from updated hospital records , careful case-control definition 878 and statistical handling of the binary outcomes , probe-level association analysis, and usage 879 of different association models to mimic various dosage mechanisms could explain the 880 increased power of our study. We consequently identified previously unreported CNV-disease 881 associations whose relevance was asserted by follow-up analyses. Only one signal – 17q12 882 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 35 CNVs increasing CKD risk – was backed by all approaches, while others were supported by 883 some analysis but not by others – e.g., two associations in the lowest statistical confidence 884 tier were replicated in the EstBB and harbored plausible candidate genes – emphasizing the 885 importance of consid ering diverse lines of corroborative evidence. Substantial overlap with 886 relevant SNP-GWAS signals and OMIM genes indicates shared genetic mechanisms with 887 convergent effects on the phenotype. Another line of evidence is coincidence with a disease-888 relevant biomarker CNV signals, e.g., four (1q21.1-2, 15q13, 16p12.2, 16p11.2 BP4-5) out of 889 six CNVRs decreasing forced vital capacity [23] were found to increase risk for pulmonary 890 diseases. This demonstrates that biomarkers are efficient proxies underlying (CNV-driven) 891 pathological processes, often increasing the statistical power to detect associations due to 892 their continuous nature. To render binary outcomes continuous, we regressed covariates out 893 of the disease status. With the same goal, more sophisticated approaches have recently been 894 developed that transform binary outcomes in continuous liability scores while borrowing 895 information from age-at-disease onset, sex, and familial history [100]. Successfully applied to 896 SNP-based GWASs, future exploration is warranted to assess their benefit in the context of 897 CNV-GWASs. By coupling a CNV -GWAS framework designed to account for challenges 898 linked to disease CNV association studies in population cohorts to extensive validation, we 899 generated a list of 73 CNV-disease pairs with various levels of supporting evidence that can 900 inform follow-up studies. 901 902 Disease-associated CNVRs harbored genes und er stronger evolutionary constraint than 903 those lacking associations and their length correlated with their propensity for pleiotropy , 904 indicating that as previously observed [8], both the number and the nature of genes affected 905 by CNVs influence their pathogenicity. Consequently, large multi -gene recurrent CNVs 906 exhibited the strongest pleiotropy . A longstanding question relates to the identification of 907 causal genes whose altered dosage drives the phenotypic alterations observed in carriers. 908 Models with various levels of complexity have been proposed, ranging from a single driver 909 gene to multiple driver genes modulated by epistatic interactions with other genes in the CNVR 910 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 36 [101]. By analyzing disease prevalence in subsets of CNV carriers, association signals could 911 be fine-mapped to narrower regions, pinpointing candidate drivers - such as ABCC6 for kidney 912 stones. In other cases, our data suggests that multiple subregions of the CNVR contribute to 913 increased risk for a given disease by cis-epistasis, as observed for 22q11.2 and ischemic 914 heart disease or 16p13.11 and hypertension. Interestingly, the putative driver for phenotypes 915 originally associated with a CNVR might not be driving our newly identified associations, as 916 shown for the 15q13 CNVR, whose non-neurological phenotypes do not appear to be linked 917 to altered dosage of CHRNA7. Beyond characterizing the pleiotropic pathological 918 consequences of recurrent CNVRs, we demonstrate that dissection of CNV -GWAS signals 919 can fine-map associations and provide mechanistic insights into their phenotypic expression. 920 921 All CNVs increased disease risk and led to an earlier age at onset. Incorporating age at onset 922 information has been shown to improve power to detect associations [100], and more 923 importantly, represents the ultimate proof of clinical relevance. Indeed, many signals mapped 924 to regions whose genetic perturbation ha s been reported to be pathogenic. These include 925 associations between well-described, clinically relevant gene-disease pairs – such as BRCA1 926 and LDLR deletions increasing the risk for ovarian cancer and IHD, respectively – but for which 927 the role of CNVs in a large population cohort had not been previously investigated. CNVs in 928 these genes have high penetrance but are extremely rare in the UKBB, so that association 929 barely reached GW significanc e. Follow-up analyses based on the medical record s, family 930 history, medication use, and biomarkers could recapitulate additional clinical associations and 931 establish that these deletions were most likely inherited, thereby generating insights into their 932 role in the general population. We further show that many CNVRs previously linked to pediatric 933 genomic disorders also increased risk for a broad spectrum of adult-onset common diseases. 934 These associations were probably overlooked as the medical consequences in adulthood of 935 these etiologies is often poorly characterized owing to ascertainment bias and difficulty to 936 gather large cohorts . While awaiting validation in clinical cohorts of CNV carriers, we hope 937 that these findings will improve clinical characterization of genomic disorders, thereby 938 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 37 facilitating diagnosis and allowing physicians to anticipate later-onset comorbidities . For 939 instance, we found carriers of 16p13.11 deletions affect ing ABCC6, the causal gene for 940 pseudoxanthoma elasticum, to be at increased risk for kidney stones, paralleling reports from 941 clinical cohorts showing that kidney stones represent an unrecognized feature of the disease 942 [90–92]. Awareness of this disease feature can mitigate kidney stone risk through adapted 943 diet and sufficient water intake. Together, our results advocate for a complex model of variable 944 CNV expressivity and penetrance that can result in a broad range of phenotypes along the 945 rare-to-common disease spectrum and represent fertile ground for in -depth, phenome-wide 946 studies aiming at better characterizing specific CNV regions [38]. 947 948 Corroborating the deleterious impact of CNVs on an individual’s health parameters, socio -949 economic status, and lifespan [18,22,23,25,31,102–105], we here speculate that it acts by 950 increasing risk for a broad range of common diseases beyond their known role in 951 neuropsychiatric disorders [4–7]. While both duplication s and deletion s contributed to 952 increased disease risk, the deletion burden’s impact was much stronger – especially for 953 metabolic, psychiatric, and musculoskeletal diseases – and for about half the diseases, 954 predominantly driven by deletion of coding region, in line with the commonly accepted view 955 that deletions tend to be more deleterious than duplications and that genetic variation affecting 956 coding regions have stronger phenotypic consequences. Only a marginal fraction of the CNV 957 burden’s contribution to disease risk was captured by disease-associated CNVRs and risk for 958 four diseases, i.e., hypothyroidism, cataract, arthrosis, and osteoporosis, associated with the 959 burden despite lacking CNV-GWAS signals. With over 10,000 cases, these diseases were not 960 underpowered compared to others, suggesting genuine differences in the genetic architecture, 961 and illustrating the added value of burden tests. Collectively, these results predict that better 962 powered CNV-GWAS are likely to isolate further associations currently captured by the CNV 963 burden. 964 965 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 38 A major limitation of our study is the reliance on microarray CNV calls, which allows to assess 966 only a fraction of the CNV landscape, i.e., mostly large CNVs or in regions with high probe 967 coverage. We speculate that small and/or multiallelic CNVs that can only be uncovered by 968 sequencing, will have a genetic architecture closer to the one of SNPs and indels, with higher 969 frequencies and more subtle effect sizes, resembling those of SNPs and indels. These effects, 970 however, are more likely tagged by common variants, limiting novel discoveries. Furthermore, 971 by detecting more events, sequencing-based studies require adapted and more stringent 972 significance thresholds. Still, having improved breakpoint resolution, such CNV calls are also 973 likely to enhance fine-mapping strategies. Microarray CNV calls also exhibit high false positive 974 rates [47]. By using stringent CNV selection criteria, we decrease the latter at the cost of 975 decreasing power to detect true associations. This aspect is particularly relevant given that 976 the type of CNVs we assess are rare and that the UKBB is depleted for disease cases [34], 977 resulting in low powered GWASs. While we adopt strategies to counter the lack of power, our 978

are likely subject to Winner’s curse, only capturing a fraction of the strongest, possibly 979 overestimated effects. This phenomenon might be compensated by UKBB CNV carriers being 980 at the milder end of the clinical spectrum, leading to effect underestimation. An interesting 981 question will be to compare effect sizes from population-based studies to those emerging from 982 clinical cohort . In the future, longitudinal follow -up of UKBB participants will increas e the 983 number of cases – especially for late -onset diseases such as Alzheimer’s or Parkinson’ 984 diseases – allowing better powered CNV -GWASs. Alternatively, meta-analyses can boost 985 case number through inclusion of clinical cohorts, at the cost of poorer disease definition due 986 to imperfect data harmonization [8]. Larger and more diverse biobanks linking genotype to 987 phenotype data [106–108] should both validate reported associations and identify new ones. 988 989

990 Our study provides in -depth analysis of the role of CNVs in modulating susceptibility to 60 991 common diseases in the general population, broadening our view on how this mutational class 992 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 39 impacts human health. Besides describing clinically relevant and actionable associations, we 993 illustrate how complex pleiotropic patterns can be dissected to gain new insights into the 994 pathological mechanisms of large recurrent CNVs, providing a framework that can be applied 995 to an even larger spectrum of diseases. 996 997 SUPPLEMENTAL DATA 998 Supplemental data include 6 figures, 6 tables, and 3 notes. 999 1000 DECLARATIONS 1001 Availability of data and materials : All data used in this study are publicly available, as 1002 described in the methods. CNV -GWAS summary statistics (UKBB) will be deposited on the 1003 GWAS Catalog upon publication and are available upon request until then. 1004 1005 Competing interests: SO is an employee of MSD at the time of the submission; contribution 1006 to the research occurred during the affiliation at the University of Lausanne. 1007 1008 Funding: The study was funded by the Swiss National Science Foundation (31003A_182632, 1009 AR; 310030_189147, ZK ), Horizon2020 Twinning projects (ePerMed 692145, AR), the 1010 Estonian Research Council (PRG687, MJ and RM), and the Department of Computational 1011 Biology (ZK) and the Center for Integrative Genomics (AR) from the University of Lausanne. 1012 1013 Author’s contributions: CA, AR and ZK conceived the study; CA carried out the analyses 1014 with contributions from MCS, NT, and CJC; The Estonian Biobank Research Team 1015 coordinated genotyping and sequencing data acquisition in the EstBB; MJ performed the 1016 replication study in the EstBB under the supervision of RM; ZK supervised statistical analyses; 1017 SO provided guidance for time -to-event analysis; CA drafted the manuscript and generated 1018 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 40 the figures; AR and ZK made critical revisions; All authors read, approved, and provided 1019 feedback on the final manuscript. 1020 1021 Acknowledgments: We thank all biobank participants for sharing their data. UKBB and 1022 EstBB computations were performed on the JURA server (University of Lausanne) and the 1023 High-Performance Computing Center (University of Tartu), respectively. 1024

1. Sudmant PH, Rausch T, Gardner EJ, Handsaker RE, Abyzov A, Huddleston J, et al. An integrated map of 1025 structural variation in 2,504 human genomes. Nature. 2015;526:75–81. 1026 2. Conrad DF, Pinto D, Redon R, Feuk L, Gokcumen O, Zhang Y, et al. Origins and fun ctional impact of copy 1027 number variation in the human genome. Nature. 2010;464:704–12. 1028 3. Zhang F, Gu W, Hurles ME, Lupski JR. Copy Number Variationin Human Health, Disease, and Evolution. Annu 1029 Rev Genomics Hum Genet. 2009;10:451–81. 1030 4. Sebat J, Lakshmi B, Malhotra D, Troge J, Lese-martin C, Walsh T, et al. Strong Association of De Novo Copy 1031 Number Mutations with Autism. Science (1979). 2007;316:445–9. 1032 5. Walsh T, McClellan JM, McCarthy SE, Addington AM, Pierce SB, Cooper GM, et al. Rare structural variants 1033 disrupt multiple genes in neurodevelopmental pathways in schizophrenia. Science (1979). 2008;320:539–43. 1034 6. Mefford HC, Muhle H, Ostertag P, von Spi czak S, Buysse K, Baker C, et al. Genome -Wide Copy Number 1035 Variation in Epilepsy: Novel Susceptibility Loci in Idiopathic Generalized and Focal Epilepsies. PLoS Genet. 1036 2010;6:e1000962. 1037 7. Cooper GM, Coe BP, Girirajan S, Rosenfeld JA, Vu TH, Baker C, et al. A copy number variation morbidity map 1038 of developmental delay. Nat Genet. 2011;43:838–46. 1039 8. Collins RL, Glessner JT, Porcu E, Lepamets M, Brandon R, Lauricella C, et al. A cross -disorder dosage 1040 sensitivity map of the human genome. Cell. 2022;185:3041-3055.e25. 1041 9. Firth H V., Richards SM, Bevan AP, Clayton S, Corpas M, Rajan D, et al. DECIPHER: Database of Chromosomal 1042 Imbalance and Phenotype in Humans Using Ensembl Resources. Am J Hum Genet. 2009;84:524–33. 1043 10. Carvalho CMB, Lupski JR. Mechanisms underlying structural variant formation in genomic disorders. Nat Rev 1044 Genet. 2016;17:224–38. 1045 11. Collins RL, Brand H, Karczewski KJ, Zhao X, Alföldi J, Francioli LC, et al. A structural variation reference for 1046 medical and population genetics. Nature. 2020;581:444–51. 1047 12. Abel HJ, Larson DE, Chiang C, Das I, Kanchi KL, Layer RM, et al. Mapping and characterization of structural 1048 variation in 17,795 deeply sequenced human genomes. Nature. 2020;583:83–9. 1049 13. Halvorsen M, Huh R, Oskolkov N, Wen J, Netotea S, Giusti-Rodriguez P, et al. Increased burden of ultra-rare 1050 structural variants localizing to boundaries of topologically associated domains in schizophrenia. Nat Commun. 1051 2020;11:1–13. 1052 14. Chen L, Abel HJ, Das I, Larson DE, Ganel L, Kanchi KL, et al. Association of structural variation with 1053 cardiometabolic traits in Finns. Am J Hum Genet. 2021;108:583–96. 1054 15. Beyter D, Ingimundardottir H, Oddsson A, Eggertsson HP, Bjornsson E, Jonsson H , et al. Long -read 1055 sequencing of 3,622 Icelanders provides insight into the role of structural variants in human diseases and other 1056 traits. Nat Genet. 2021;53:779–86. 1057 16. Babadi M, Fu JM, Lee SK, Smirnov AN, Gauthier LD, Walker M, et al. GATK -gCNV: A Rare Copy Number 1058 Variant Discovery Algorithm and Its Application to Exome Sequencing in the UK Biobank. bioRxiv. 1059 2022;2022.08.25.504851. 1060 17. Fitzgerald T, Birney E. CNest: A novel copy number association discovery method uncovers 862 new 1061 associations from 200 ,629 whole -exome sequence datasets in the UK Biobank. Cell Genomics. 1062 2022;2:100167. 1063 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 41 18. Kendall KM, Rees E, Escott -Price V, Einon M, Thomas R, Hewitt J, et al. Cognitive Performance Among 1064 Carriers of Pathogenic Copy Number Variants: Analysis of 152,000 UK Biobank Subjects. Biol Psychiatry. 1065 2017;82:103–10. 1066 19. Owen D, Bracher-Smith M, Kendall KM, Rees E, Einon M, Escott-Price V, et al. Effects of pathogenic CNVs on 1067 physical traits in participants of the UK Biobank. BMC Genomics. 2018;19:1–9. 1068 20. Crawford K, Bracher-Smith M, Owen D, Kendall KM, Rees E, Pardiñas AF, et al. Medical consequences of 1069 pathogenic CNVs in adults: Analysis of the UK Biobank. J Med Genet. 2019;56:131–8. 1070 21. Kendall KM, Rees E, Bracher-Smith M, Legge S, Riglin L, Zammit S, et al. Association of Rare Copy Number 1071 Variants with Risk of Depression. JAMA Psychiatry. 2019;76:818–25. 1072 22. Macé A, Tuke MA, Deelen P, Kristiansson K, Mattsson H, Nõukas M, et al. CNV-association meta-analysis in 1073 191,161 European adults reveals new loci associated with anthropometric traits. Nat Commun. 2017;8:1–11. 1074 23. Auwerx C, Lepamets M, Sadler MC, Patxot M, Stojanov M, Baud D, et al. The individual and global impact of 1075 copy-number variants on complex human traits. Am J Hum Genet. 2022;109:647–68. 1076 24. Aguirre M, Rivas MA, Priest J. Phenome-wide Burden of Copy-Number Variation in the UK Biobank. Am J Hum 1077 Genet. 2019;105:373–83. 1078 25. Hujoel MLA, Sherman MA, Barton AR, Mukamel RE, Sankaran VG, Terao C, et al. Influences of rare copy -1079 number variation on human complex traits. Cell. 2022;185:4233-4248.e27. 1080 26. Sinnott-Armstrong N, Tanigawa Y, Amar D, Mars N, Benner C, Aguirre M, et al. Genetics of 35 blood and urine 1081 biomarkers in the UK Biobank. Nat Genet. 2021;53:185–94. 1082 27. Li YR, Glessner JT, Coe BP, Li J, Mohebnasab M, Chang X, et al. Rare copy number variants in over 100,000 1083 European ancestry subjects reveal multiple disease associations. Nat Commun. 2020;11:1–9. 1084 28. Kopal J, Kumar K, Saltoun K, Modenato C, Moreau CA, Martin-Brevet S, et al. Rare CNVs and phenome-wide 1085 profiling highlight brain structural divergence and phenotypical convergence. Nat Hum Behav. 2023; 1086 29. Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep 1087 phenotyping and genomic data. Nature. 2018;562:203–9. 1088 30. Wright CF, West B, Tuke M, Jones SE, Patel K, Laver TW, et al. Assessing the Pathogenicity, Penetrance, and 1089 Expressivity of Putative Disease-Causing Variants in a Population Setting. Am J Hum Genet. 2019;104:275–1090 86. 1091 31. Kingdom R, Tuke M, Wood A, Beaumont RN, Frayling TM, Weedon MN, et al. Rare genetic variants in genes 1092 and loci linked to dominant monogenic developmental disorders cause milder related phenotypes in the general 1093 population. Am J Hum Genet. 2022;109:1308–16. 1094 32. Chen R, Shi L, Hakenberg J, Naughton B, Sklar P, Zhang J, et al. Analysis of 589,306 genomes identifies 1095 individuals resilient to severe Mendelian childhood diseases. Nat Biotechnol. 2016;34:531–8. 1096 33. Goodrich JK, Singer-Berk M, Son R, Sveden A, Wood J, England E, et al. Determinants of penetrance and 1097 variable expressivity in monogenic metabolic conditions across 77,184 exomes. Nat Commun. 2021;12:1–15. 1098 34. Fry A, Littlejohns TJ, Sudlow C, Doherty N, Adamska L, Sprosen T, et al. Comparison of Sociodem ographic 1099 and Health-Related Characteristics of UK Biobank Participants with Those of the General Population. Am J 1100 Epidemiol. 2017;186:1026–34. 1101 35. Falconer D. The inheritance of liability to certain diseases, estimated from the incidence among relatives. Ann 1102 Hum Genet. 1965;29:51–76. 1103 36. Senn StephenS. Statistical issues in drug development. Wiley; 2021. 1104 37. Mollon J, Schultz LM, Huguet G, Knowles EEM, Mathias SR, Rodrigue A, et al. Impact of Copy Number Variants 1105 and Polygenic Risk Scores on Psychopathology in the UK Biobank. Biol Psychiatry. 2023; 1106 38. Zamariolli M, Auwerx C, Sadler MC, Graaf A Van Der, Lepik K, Schoeler T, et al. The impact of 22q11 . 2 copy-1107 number variants on human traits in the general population. Am J Hum Genet. 2023;110:1–14. 1108 39. Giannuzzi G, Schmidt PJ, Porcu E, Willemin G, Munson KM, Nuttle X, et al. The Human -Specific BOLA2 1109 Duplication Modifies Iron Homeostasis and Anemia Predisposition in Chromosome 16p11.2 Autism Individuals. 1110 Am J Hum Genet. 2019;105:947–58. 1111 40. Leitsalu L, Haller T, Esko T, Tammesoo ML, Alavere H, Snieder H, et al. Cohort profile: Estonian biobank of 1112 the Estonian genome center, university of Tartu. Int J Epidemiol. 2015;44:1137–47. 1113 41. Wu P, Gifford A, Meng X, Li X, Campbell H, Varley T, et al. Mapping ICD -10 and ICD -10-CM Codes to 1114 Phecodes: Workflow Development and Initial Evaluation. JMIR Med Inform. 2019;7:e14325. 1115 42. Sollis E, Mosaku A, Abid A, Buniello A, Cerezo M, Gil L, et al. The NHGRI-EBI GWAS Catalog: knowledgebase 1116 and deposition resource. Nucleic Acids Res. 2023;51:D977–85. 1117 43. Hamosh A, Scott AF, Amberger JS, Bocchini CA, McKusick VA. Online Mendelian Inheritance in Man (OMIM), 1118 a knowledgebase of human genes and genetic disorders. Nucleic Acids Res. 2005;33. 1119 44. Karczewski KJ., Francioli LC, Tiao G, Cummings BB, Alföldi J, Wang Q, et al. The mutational constraint 1120 spectrum quantified from variation in 141,456 humans. Nature. 2020;581:434–43. 1121 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 42 45. Aguet F, Barbeira A, Bona zzola R, Brown A, Castel S, Jo B, et al. The GTEx Consortium atlas of genetic 1122 regulatory effects across human tissues. Science (1979). 2020;369:1318–30. 1123 46. Wang K, Li M, Hadley D, Liu R, Glessner J, Grant SFA, et al. PennCNV: An integrated hidden Markov model 1124 designed for high-resolution copy number variation detection in whole-genome SNP genotyping data. Genome 1125 Res. 2007;17:1665–74. 1126 47. Macé A, Tuke MA, Beckmann JS, Lin L, Jacquemont S, Weedon MN, et al. New quality measure for SNP array 1127 based CNV detection. Bioinformatics. 2016;32:3298–305. 1128 48. Chang CC, Chow CC, Tellier LCAM, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: Rising to the 1129 challenge of larger and richer datasets. Gigascience. 2015;4:1–16. 1130 49. Wang K, Li M, Hakonarson H. ANNOVA R: functional annotation of genetic variants from high -throughput 1131 sequencing data. Nucleic Acids Res. 2010;38:e164–e164. 1132 50. Kent WJ, Sugnet CW, Furey TS, Roskin KM, Pringle TH, Zahler AM, et al. The human genome browser at 1133 UCSC. Genome Res. 2002;12:996–1006. 1134 51. Gao X, Starmer J, Martin ER. A multiple testing correction method for genetic association studies using 1135 correlated single nucleotide polymorphisms. Genet Epidemiol. 2008;32:361–9. 1136 52. Therneau TM. Survival Analysis [R package survival version 3.5-3]. 2022 [cited 2023 Apr 17]; Available from: 1137 https://CRAN.R-project.org/package=survival 1138 53. Berman JJ. Rare Diseases and Orphan Drugs: Keys to Understanding and Treating the Common Diseases. 1139 Elsevier Inc.; 2014. 1140 54. Li A, Jiao X, Munier FL, Schorderet DF, Yao W, Iwata F, et al. Bietti Crystalline Corneoretinal Dystrophy Is 1141 Caused by Mutations in the Novel Gene CYP4V2. Am J Hum Genet. 2004;74:817–26. 1142 55. Zhou W, Otto EA, Cluckey A, Airik R, Hurd TW, Chaki M, et al. FAN1 mutations cause karyomegalic interstitial 1143 nephritis, linking chronic kidney failure to defective DNA damage repair. Nat Genet. 2012;44:910–5. 1144 56. Kryger MH, Roth T, Goldstein CA. Principles and Practice of Sleep Medicine. Elsevier Health Sciences; 2021. 1145 57. Shinawi M, Liu P, Kang SHL, Shen J, Belmont JW, Scott DA, et al. Recurrent reciprocal 16p11.2 1146 rearrangements associated with global developmental delay, behavioural problems, dysmorphism, epilepsy, 1147 and abnormal head size. J Med Genet. 2010;47:332–41. 1148 58. Weiss LA, Shen Y, Korn JM, Arking DE, Miller DT, Fossdal R, et al. Association between Microdeletion and 1149 Microduplication at 16p11.2 and Autism. N Engl J Med. 2008;358:667–75. 1150 59. D’Angelo D, Lebon S, Chen Q, Martin -Brevet S, Snyder LAG, Hippolyte L, et al. Defining the Effect of the 1151 16p11.2 Duplication on Cognition, Behavior, and Medical Comorbidities. JAMA Psychiatry. 2016;73:20–30. 1152 60. Reinthaler EM, Lal D, Lebon S, Hildebra nd MS, Dahl HHM, Regan BM, et al. 16p11.2 600 kb Duplications 1153 confer risk for typical and atypical Rolandic epilepsy. Hum Mol Genet. 2014;23:6069–80. 1154 61. Jacquemont S, Reymond A, Zufferey F, Harewood L, Walters RG, Kutalik Z, et al. Mirror extreme BMI 1155 phenotypes associated with gene dosage at the chromosome 16p11.2 locus. Nature. 2011;478:97–102. 1156 62. McCarthy SE, Makarov V, Kirov G, Addington AM, McClellan J, Yoon S, et al. Microduplications of 16p11.2 are 1157 associated with schizophrenia. Nat Genet. 2009;41:1223–7. 1158 63. Walters RG, Jacquemont S, Valsesia A, De Smith AJ, Martinet D, Andersson J, et al. A new highly penetrant 1159 form of obesity due to deletions on chromosome 16p11.2. Nature. 2010;463:671–5. 1160 64. Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S, et al. A strong candidate for the 1161 breast and ovarian cancer susceptibility gene BRCA1. Science (1979). 1994;266:66–71. 1162 65. Iacocca MA, Hegele RA. Role of DNA copy number variation in dyslipidemias. Curr Opin Lipidol. 2018;29:125–1163 32. 1164 66. Hobbs HH, Russell DW, Brown MS, Goldstein JL. The LDL receptor locus in familial hypercholesterolemia: 1165 mutational analysis of a membrane protein. Annu Rev Genet. 1990;24:133–70. 1166 67. Defesche JC, Gidding SS, Harada -Shiba M, Hegele RA, Santos RD, Wie rzbicki AS. Familial 1167 hypercholesterolaemia. Nat Rev Dis Primers. 2017;3:17093. 1168 68. Iacocca MA, Wang J, Dron JS, Robinson JF, McIntyre AD, Cao H, et al. Use of next-generation sequencing to 1169 detect LDLR gene copy number variation in familial hypercholesterolemia. J Lipid Res. 2017;58:2202–9. 1170 69. Mach F, Baigent C, Catapano AL, Koskina s KC, Casula M, Badimon L, et al. 2019 ESC/EAS Guidelines for 1171 the management of dyslipidaemias: lipid modification to reduce cardiovascular riskThe Task Force for the 1172 management of dyslipidaemias of the European Society of Cardiology (ESC) and European Ath erosclerosis 1173 Society (EAS). Eur Heart J. 2020;41:111–88. 1174 70. Wuttke M, Li Y, Li M, Sieber KB, Feitosa MF, Gorski M, et al. A catalog of genetic loci associated with kidney 1175 function from analyses of a million individuals. Nat Genet. 2019;51:957–72. 1176 71. Stanzick KJ, Li Y, Schlosser P, Gorski M, Wuttke M, Thomas LF, et al. Discovery and prioritization of variants 1177 and genes for kidney function in >1.2 million individuals. Nat Commun. 2021;12:1–17. 1178 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 43 72. Fajans SS, Bell GI, Polonsky KS. Molecular mechanisms and clinical pathophysiology of maturity -onset 1179 diabetes of the young. N Engl J Med. 2001;345:97180. 1180 73. Mefford HC, Clauin S, Sharp AJ, Moller RS, Ullmann R, Kapur R, et al. Recurrent Reciprocal Genomic 1181 Rearrangements of 17q12 Are Associated with Renal Disea se, Diabetes, and Epilepsy. Am J Hum Genet. 1182 2007;81:1057–69. 1183 74. Girirajan S, Rosenfeld JA, Cooper GM, Antonacci F, Siswara P, Itsara A, et al. A recurrent 16p12.1 1184 microdeletion supports a two-hit model for severe developmental delay. Nat Genet. 2010;42:203–9. 1185 75. Stefansson H, Meyer -Lindenberg A, Steinberg S, Magnusdottir B, Morgen K, Arnarsdottir S, et al. CNVs 1186 conferring risk of autism or schizophrenia affect cognition in controls. Nature. 2013;505:361–6. 1187 76. Girirajan S, Pizzo L, Moeschler J, Rosenfeld J. 16p12.2 Recurrent Deletion. GeneReviews® [Internet]. 2018 1188 [cited 2023 Apr 4]; Available from: https://www.ncbi.nlm.nih.gov/books/NBK274565/ 1189 77. Stevelink R, Campbell C, Chen S, The International League Against Epilepsy Consortium on Complex 1190 Epilepsies. Genome-wide meta-analysis of over 29,000 people with epilepsy reveals 26 loci and subtype -1191 specific genetic architecture. medRxiv. 2022;2022.06.08.22276120. 1192 78. Howles SA, Wiberg A, Goldsworthy M, Bayliss AL, Gluck AK, Ng M, et al. Genetic variants of calcium and 1193 vitamin D metabolism in kidney stone disease. Nat Commun. 2019;10:1–10. 1194 79. Evangelou E, Warren HR, Mosen-Ansorena D, Mifsud B, Pazoki R, Gao H, et al. Genetic analysis of over 1 1195 million people identifies 535 new loci associated with blood pressure traits. Nat Genet. 2018;50:1412–25. 1196 80. De Kovel CGF, Trucks H, Helbig I, Mefford HC, Baker C, Leu C, et al. Recurrent microdeletions at 15q11.2 and 1197 16p13.11 predispose to idiopathic generalized epilepsies. Brain. 2010;133:23–32. 1198 81. Heinzen EL, Radtke RA, Urban TJ, Cavalleri GL, Depondt C, Need AC, et al. Rare Deletions at 16p13.11 1199 Predispose to a Diverse Spectrum of Sporadic Epilepsy Syndromes. Am J Hum Genet. 2010;86:707–18. 1200 82. Alkuraya FS, Cai X, Emery C, Mochida GH, Al-Dosari MS, Felie JM, et al. Human Mutations in NDE1 Cause 1201 Extreme Microcephaly with Lissencephaly. Am J Hum Genet. 2011;88:536–47. 1202 83. Bakircioglu M, Carvalho OP, Khurshid M, Cox JJ, Tuysuz B, Barak T, et al. The Essential Role of Centrosomal 1203 NDE1 in Human Cerebral Cortex Neurogenesis. Am J Hum Genet. 2011;88:523–35. 1204 84. Ringpfeil F, Lebwohl MG, Christiano AM, Uitto J. Pseudoxanthoma elasticum: Mutations in the MRP6 gene 1205 encoding a transmembrane ATP-binding cassette (ABC) transporter. Proc Natl Acad Sci U S A. 2000;97:6001–1206 6. 1207 85. Struk B, Cai L, Zäch S, Ji W, Chung J, Lumsden A, et al. Mutations of the gene encoding the transmembrane 1208 transporter protein ABC-C6 cause pseudoxanthoma elasticum. J Mol Med. 2000;78:282–6. 1209 86. Le Saux O, Urban Z, Tschuch C, Csiszar K, Bacchelli B, Quaglino D, et al. Mutations in a gene encoding an 1210 ABC transporter cause pseudoxanthoma elasticum. Nat Genet. 2000;25:223–7. 1211 87. Bergen AAB, Plomp AS, Schuurman EJ, Terry S, Breuning M, Dauwerse H, et al. Mutations in ABCC6 cause 1212 pseudoxanthoma elasticum. Nat Genet. 2000;25:228–31. 1213 88. Le Saux O, Beck K, Sachsinger C, Silvestri C, Treiber C, Goöring HHH, et al. A Spectrum of ABCC6 Mutations 1214 Is Responsible for Pseudoxanthoma Elasticum. Am J Hum Genet. 2001;69:749–64. 1215 89. Ringpfeil F, Nakano A, Uitto J, Pulkkinen L. Compound Heterozygosity for a Recurrent 16.5-kb Alu-Mediated 1216 Deletion Mutation and Single -Base-Pair Substitutions in the ABCC6 Gene Results in Pseudoxanthoma 1217 Elasticum. Am J Hum Genet. 2001;68:642–52. 1218 90. Ralph D, Allawh R, Terry IF, Terry SF, Uitto J, Li QL. Kidney Stones Are Prevalent in Individuals with 1219 Pseudoxanthoma Elasticum, a Genetic Ectopic Mineralization Disorder. Int J Dermatol Venereol. 2020;3:198–1220 204. 1221 91. Legrand A, Cornez L, Samkari W, Mazzella JM, Venisse A, Boccio V, et al. Mutation spectrum in the ABCC6 1222 gene and genotype–phenotype correlations in a French cohort with pseudoxanthoma elasticum. Genetics in 1223 Medicine. 2017;19:909–17. 1224 92. Letavernier E, Kauffenstein G, Hugue t L, Navasiolava N, Bouderlique E, Tang E, et al. ABCC6 deficiency 1225 promotes development of randall plaque. Journal of the American Society of Nephrology. 2018;29:2337–47. 1226 93. Nitschke Y, Baujat G, Botschen U, Wittkampf T, Du Moulin M, Stella J, et al. Generalized Arterial Calcification 1227 of Infancy and Pseudoxanthoma Elasticum Can Be Caused by Mutations in Either ENPP1 or ABCC6. Am J 1228 Hum Genet. 2012;90:25–39. 1229 94. Le Boulanger G, Labrèze C, Croué A, Schurgers LJ, Chassaing N, Wittkampf T, et al. An unusual s evere 1230 vascular case of pseudoxanthoma elasticum presenting as generalized arterial calcification of infancy. Am J 1231 Med Genet A. 2010;152A:118–23. 1232 95. McDonald-McGinn DM, Sullivan KE, Marino B, Philip N, Swillen A, Vorstman JAS, et al. 22q11.2 deletion 1233 syndrome. Nat Rev Dis Primers. 2015;1:1–19. 1234 96. Bartik LE, Hughes SS, Tracy M, Feldt MM, Zhang L, Arganbright J, et al. 22q11.2 duplications: Expanding the 1235 clinical presentation. Am J Med Genet A. 2022;188:779–87. 1236 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint Auwerx et al., 2023 44 97. Sharp AJ, Mefford HC, Li K, Baker C, Skinner C, Stevenson RE, et al. A recurrent 15q13.3 microdeletion 1237 syndrome associated with mental retardation and seizures. Nat Genet. 2008;40:322–8. 1238 98. Lowther C, Costain G, Stavropoulos DJ, Melvin R, Silversides CK, Andrade DM, et al. Delineating the 15q13.3 1239 microdeletion phenotype: a case series and comprehensive review of the literature. Genetics in Medicine. 1240 2015;17:149–57. 1241 99. Gillentine MA, S chaaf CP. The Human Clinical Phenotypes of Altered CHRNA7 Copy Number. Biochem 1242 Pharmacol. 2015;97:352. 1243 100. Pedersen EM, Agerbo E, Plana-Ripoll O, Grove J, Dreier JW, Musliner KL, et al. Accounting for age of onset 1244 and family history improves power in genome-wide association studies. Am J Hum Genet. 2022;109:417–32. 1245 101. Golzio C, Katsanis N. Genetic architecture of reciprocal CNVs. Curr Opin Genet Dev. 2013;23:240–8. 1246 102. Männik K, Mägi R, Macé A, Cole B, Guyatt AL, Shihab HA, et al. Copy number variati ons and cognitive 1247 phenotypes in unselected populations. JAMA. 2015;313:2044–54. 1248 103. Wheeler E, Huang N, Bochukova EG, Keogh JM, Lindsay S, Garg S, et al. Genome -wide SNP and CNV 1249 analysis identifies common and low-frequency variants associated with severe early-onset obesity. Nat Genet. 1250 2013;45:513–7. 1251 104. Dauber A, Yu Y, Turchin MC, Chiang CW, Meng YA, Demerath EW, et al. Genome-wide association of copy-1252 number variation reveals an association between short stature and the presence of low -frequency genomic 1253 deletions. Am J Hum Genet. 2011;89:751–9. 1254 105. Saarentaus EC, Havulinna AS, Mars N, Ahola-Olli A, Kiiskinen TTJ, Partanen J, et al. Polygenic burden has 1255 broader impact on health, cognition, and socioeconomic outcomes than most rare and high-risk copy number 1256 variants. Mol Psychiatry. 2021;26:4884–95. 1257 106. Denny JC, Rutter JL, Goldstein DB, Philippakis A, Smoller JW, Jenkins G, et al. The ‘All of Us’ Research 1258 Program. N Engl J Med. 2019;381:668–76. 1259 107. Hunter-Zinck H, Shi Y, Li M, Gorman BR, Ji SG, Sun N, et al. Genotyping Array Design and Data Qualit y 1260 Control in the Million Veteran Program. Am J Hum Genet. 2020;106:535–48. 1261 108. Kurki MI, Karjalainen J, Palta P, Sipilä TP, Kristiansson K, Donner KM, et al. FinnGen provides genetic insights 1262 from a well-phenotyped isolated population. Nature. 2023;613:508–18. 1263 . CC-BY 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted August 5, 2023. ; https://doi.org/10.1101/2023.07.31.23293408doi: medRxiv preprint