{"paper_id":"33dc264c-83bb-49a5-9768-74cf166d2938","body_text":"InteracTor: Feature Engineering and Explainable AI for Profiling Protein Structure-Interaction-Function Relationships | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return\"[object Function]\"==o.call(a)}function e(a){return\"string\"==typeof a}function f(){}function g(a){return!a||\"loaded\"==a||\"complete\"==a||\"uninitialized\"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){(\"c\"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){\"img\"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),\"object\"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height=\"0\",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),\"img\"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||\"j\",e(a)?i(\"c\"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName(\"script\")[0],o={}.toString,p=[],q=0,r=\"MozAppearance\"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&\"[object Opera]\"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?\"object\":l?\"script\":\"img\",v=l?\"script\":u,w=Array.isArray||function(a){return\"[object Array]\"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split(\"!\"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split(\"=\"),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(\".\").pop().split(\"?\").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split(\"/\").pop().split(\"?\")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&\"css\"==i.url.split(\".\").pop().split(\"?\").shift()?\"c\":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results InteracTor: Feature Engineering and Explainable AI for Profiling Protein Structure-Interaction-Function Relationships View ORCID Profile Jose Cleydson F. Silva , Layla Schuster , Nick Sexson , Melissa Erdem , Ryan Hulk , Matias Kirst , Marcio F. R. Resende , View ORCID Profile Raquel Dias doi: https://doi.org/10.1101/2025.04.10.648139 Jose Cleydson F. Silva 1 Department of Microbiology and Cell Science, Institute of Food and Agricultural Sciences, University of Florida , Gainesville, FL, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jose Cleydson F. Silva Layla Schuster 1 Department of Microbiology and Cell Science, Institute of Food and Agricultural Sciences, University of Florida , Gainesville, FL, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Nick Sexson 1 Department of Microbiology and Cell Science, Institute of Food and Agricultural Sciences, University of Florida , Gainesville, FL, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Melissa Erdem 1 Department of Microbiology and Cell Science, Institute of Food and Agricultural Sciences, University of Florida , Gainesville, FL, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Ryan Hulk 1 Department of Microbiology and Cell Science, Institute of Food and Agricultural Sciences, University of Florida , Gainesville, FL, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Matias Kirst 3 Genetics Institute, University of Florida , Gainesville, FL, USA 4 School of Forest, Fisheries and Geomatic Sciences, University of Florida , Gainesville, FL, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Marcio F. R. Resende 2 Horticultural Sciences Department, University of Florida , Gainesville, FL, United States Find this author on Google Scholar Find this author on PubMed Search for this author on this site Raquel Dias 1 Department of Microbiology and Cell Science, Institute of Food and Agricultural Sciences, University of Florida , Gainesville, FL, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Raquel Dias For correspondence: raquel.dias{at}ufl.edu Abstract Full Text Info/History Metrics Preview PDF ABSTRACT Characterizing protein families’ structural and functional diversity is essential for understanding their biological roles. Traditional analyses often focus on primary and secondary structures, which may not fully capture complex protein interactions. Here we introduce InteracTor, a novel toolkit that extracts multimodal features from protein three-dimensional (3D) structures, including interatomic interactions like hydrogen bonds, van der Waals forces, and hydrophobic contacts. By integrating Explainable AI (XAI) techniques, we quantified the importance of the extracted features in the classification of protein structural and functional families. InteracTor’s interpretable features enable mechanistic insights into the determinants of protein structure, function, and dynamics, offering a transparent means to assess their predictive power within machine learning models. Interatomic interaction features extracted by InteracTor demonstrated superior predictive power for protein family classification compared to features based solely on primary or secondary structure, revealing the importance of considering specific tertiary contacts in computational protein analysis. This work provides a robust framework for future studies aiming to enhance the capabilities of models for protein function prediction and drug discovery. AUTHOR SUMMARY InteracTor is a computational toolkit designed to enhance our understanding of protein structure and function by focusing on three-dimensional (3D) structural interactions. Unlike traditional approaches that primarily rely on sequence or secondary structure data, InteracTor extracts biologically meaningful features such as hydrogen bonds, van der Waals forces, and hydrophobic contacts, which are critical for protein stability and dynamics. By integrating these features into machine learning models alongside explainable AI methods, InteracTor provides interpretable insights into how specific structural interactions influence protein behavior. Our results demonstrate that tertiary structure features significantly improve the accuracy of protein family classification compared to sequence-based methods alone, underscoring the importance of considering 3D interactions in computational protein analyses. The toolkit’s modular design makes it adaptable for diverse applications, including drug discovery and protein engineering. In a broader context, InteracTor bridges the gap between computational biology and practical applications in medicine and biotechnology by offering a transparent and robust framework for analyzing proteins at a molecular level. This work represents a step forward in leveraging structural data to advance predictive modeling and biological discovery. INTRODUCTION In recent decades, high-throughput sequencing techniques have dramatically expanded protein sequence databases. At the same time, advances in cryo-electron microscopy and deep-learning-based computational structure determination methods, including AlphaFold( 1 ) and RoseTTAFold( 2 ), have transformed protein structure elucidation. Consequently, the surge in available sequence and structural data has catalyzed the development of machine- and deep-learning techniques for predictive modeling. This data has been leveraged to address a variety of challenges, such as identifying non-classical secreted proteins( 3 ),( 4 ) predicting binding affinity( 5 ),( 6 ),( 7 ), and engineering proteins for novel functions( 8 , 9 ). Central to these algorithms is feature engineering and encoding, aimed at converting protein sequences and physiochemical properties into machine-readable formats. Ideally, this process captures the attributes most relevant to the predictive targets of interest. Sequence-based feature representations are among the most widely utilized, including amino acid composition, chemical property-based features, k-mers, and alignment-based embeddings. These descriptors effectively simplify sequence information and reduce the data dimensionality while still highlighting broader functional characteristics, sequence patterns, and evolutionary relationships. However, sequence-based methods can suffer from high dimensionality and data sparsity and are limited in their ability to capture critical properties influencing protein function. The incorporation of 3D structural data into the suite of available encodings allows predictive models to have a deeper layer of biological context that can give insight into functional dynamics. This has spurred the development of more comprehensive feature extraction platforms such as iFeatureOmega( 10 ), Pfeature( 11 ), and ProFeatX( 12 ) which incorporate both sequence and secondary structure descriptors. Pfeature utilizes amino acid sequences, employing binary encoding for chemical elements and leveraging PaDEL software for fingerprint generation. ProFeatX focuses on torsional angle bigrams, providing insights into secondary structure. This diverse array of methodologies underscores the complementary nature of these tools, with each excelling in specific bioinformatics applications. However, these tools do not consider the three-dimensional structure of the protein, nor do they consider the interactions between amino acids within the protein’s three-dimensional framework. Tools like iFeature harness molecular structures to calculate Half Sphere Exposure (HSE) and Solvent Accessible Surface (SAS), but still lack support for the extraction of interatomic interactions and other key structural features( 10 ). While traditional protein sequence descriptors have been fundamental for many predictive modeling types, encoding interaction features remains a significant yet underutilized strategy for capturing the nuances of protein behavior. Although the recent surge in powerful sequence and structural embeddings that significantly improves predictive performance, they often create uninterpretable ’black-box’ models, making it extremely difficult to understand the relationship between the input features and model predictions, even with advanced Explainable AI (XAI) techniques( 13 ). In contrast, InteracTor’s extracted interatomic interaction features are inherently biologically meaningful, allowing for more direct and transparent application of XAI in downstream models. This enables researchers to pinpoint specific interatomic interaction types responsible for observed protein structural and functional properties, offering a clearer path to biological insight. Here we present InteracTor, a toolkit for the extraction of three types of protein feature encodings: interaction features, physicochemical features, and compositional features. Interaction features include hydrogen bonds, hydrophobic contacts, repulsive interactions, and van der Waals interactions, each encoding unique aspects of molecular dynamics that play an important role in governing protein function. Specifically, hydrogen bonds and hydrophobic contacts are important for stabilizing secondary and tertiary structures( 14 ),( 15 ). Van der Waals interactions influence molecular complementarity, which is crucial for substrate binding, and mediate transient interactions that can facilitate or destabilize protein structures and complexes( 16 ). Physiochemical property features include accessible solvent area, hydrophobicity, and surface tension, which are implicated in protein folding, stability, solubility, and protein-protein interactions. Compositional features include mono-, di-, and tripeptide composition and amino acid side chain chemical property (CPAASC) frequencies( 17 ),( 18 ),( 19 ). These features determine local spatial arrangements (secondary structure) and the overall 3D folded conformation (tertiary structure) of a protein through the formation of alpha helices, beta sheets, loops, and structural motifs. By leveraging XAI techniques such as Shapley Additive exPlanations (SHAP)( 20 ), we compared the importance of InteracTor’s extracted interatomic interaction features to classic primary and secondary structure features across multiple machine learning model architectures. Our feature sets directly map to biologically meaningful concepts, enabling users to readily interpret results and validate the logic of explainable AI models, in contrast to abstract embedding or principal component vectors. This approach allowed us to pinpoint the most impactful features for characterizing protein families and elucidate the relative contribution of tertiary structural information to predictive performance. Structural and sequence-based features were complementary and provided a more comprehensive representation of the protein. Our integrative feature encoding and selection approach underscores the complexity and richness of proteins, ultimately advancing our ability to characterize proteins for various applications in structural biology, biotechnology and medicine. RESULTS Multimodal protein profiling InteracTor computes 11 different interatomic interaction features and 8 distinct CPAASC that are key for characterizing structure and function of proteins ( Table 1 and Supplementary Table 1 ). In addition to interaction and structural features, our toolkit also extracts classic protein sequence compositional features such as k-mer frequencies ( Supplementary Table 2 ), resulting in a total of 18,296 multimodal features extracted ( Table 2 ). This provides a comprehensive representation of protein structure, function, and sequence characteristics, enabling in-depth analysis of protein properties across various scales of molecular organization. While we demonstrate InteracTor’s utility through protein function family classification, the toolkit is designed for modular adaptation to diverse structural biology tasks—from drug binding analysis to protein engineering—thanks to its interpretable, feature-driven framework. View this table: View inline View popup Table 1: Protein structural features based interatomic interactions. View this table: View inline View popup Table 2: Overview of protein features extracted. InteracTor characterizes the variability among distinct protein families We conducted PCA across all 18,296 features extracted from 20,877 protein structures representing the most abundant protein families and GO terms in the PDB REDO database ( Supplementary Tables 3-4 ) to evaluate the overall utility of InteracTor in the characterization of variability among protein families and Gene Ontology (GO) terms. Fig. 2 shows the first two of 500 selected PCAs ( Supplementary Fig. S1C-D ), which captured 1.72% of the variance across protein families ( Fig. 2B ), and 1.0% of the variance across GO terms ( Fig. 2A ). Except for Peptidase S1 and the Glycosyl hydrolase 5 (cellulase A), the protein families exhibited well-defined clusters in the two-dimensional space ( Fig. 2D ), whereas GO terms exhibited less separation overall ( Fig. 2A ). This difference may be attributed to the inconsistent accuracy of GO annotations across non-model organisms( 31 ). Download figure Open in new tab Fig. 1. Pipeline of the InteracTor Algorithm. (A) Extensive database of PDB files, where each chain is split into an individual PDB file and converted to MOL2 format. (B) The following metrics are calculated: atom-atom distances (d) and van der Waals radii (r); bond angles (a); geometric centers (g); interaction types and atom types. Molecular representation with carbon atoms in gray, oxygen in red, and nitrogen in blue. (C) The following features are calculated: interatomic interaction features ( Table 1 ); side chain physicochemical properties ( Supplementary Table S1 ); mono-, di-, and tripeptide composition ( Supplementary Table S1 ). (D) The extracted features are exported in tabular format, which then is utilized for feature selection, dimensionality reduction, clustering, visualization, and machine learning. Download figure Open in new tab Fig. 2. Analysis of Protein Families and Their Molecular Functions. A) Clustering of three representative groups based on molecular functions. B) Clustering of the most prominent protein family, showcasing distinct traits unique to that family. C) Three-dimensional PCA visualization of molecular functions, capturing the variance and complexity within different functional categories. D) Three-dimensional PCA representation of protein families. Three-dimensional projections provided clearer clustering of the data, allowing a distinct visualization of relationships among the protein families. The first three principal components captured 2.40% of the variance across protein families ( Fig. 2C , Fig. S1A ), as well as 1.38% of the variance across GO terms ( Fig. 2D , Fig. S1B ). The 3D visualization also improved the separation of Peptidase S1 and Glycosyl hydrolase 5 families, as well as the separation of GO groups, allowing for a more nuanced interpretation of functional relationships among groups ( Fig. 2C ). Mutual Information scoring identifies key features across protein families To enhance clustering quality and classification accuracy in downstream analyses, InteracTor performs feature selection using Mutual Information (MI) scoring. All features were ranked by MI score to prioritize those most relevant to the target variables. Fig. 3A shows the distribution of MI scores used for feature selection. The distribution exhibits a primary mode, or large peak, which corresponds to features with low MI scores (MI<0.2), reflecting background noise and less informative features. The remaining lower peaks (MI>=0.2) represent a subset of 354 features with informational content effectively distinguished from the background noise peak. Among the top 100 high MI scoring features are 9 interatomic interactions, 2 CPAASC features, and 89 sequence composition features ( Supplementary Table 5 ). Among the 12 most highly ranked features across protein families ( Fig. 3B ) are hydrogen bonds (MI=0.775), total surface tension (MI=0.763), London dispersion forces (MI=0.758), repulsive interactions (MI=0.722), internal tension (MI=0.708), Accessible Surface Area (ASA) (MI=0.694), hydrophobic contacts (MI=0.561), TG frequency (MI=0.562), internal hydrophobicity (MI=0.561), VN frequency (MI=0.556), total hydrophobicity (MI=0.539), and GG frequency (MI=0.509). Download figure Open in new tab Fig. 3. Mutual Information (MI) Scores Across Protein Families and Molecular Functions. A) Frequency distribution of MI scores for each protein family B) Density distribution of MI scores across protein families, focusing on the top 12 ranked features. C) Frequency distribution of MI scores for molecular functions. D) Density distribution of MI scores for molecular functions, focusing on the top 12 ranked features. For GO terms, the MI score analysis revealed values ranging from approximately 0.06 to 0.1, as shown in Fig. 3C . While several peaks are well-defined, indicating key features, the MI scores are notably lower compared to those for protein families. Fig. 3D presents a violin plot illustrating the distribution of MI scores, with a high density of values between 0.1 and 0.2. High MI scoring features include total surface tension (MI=0.202), London dispersion forces (MI=0.165), total hydrophobicity (MI=0.146), ASA (MI=0.142), EL frequence (MI=0.127), VDW interactions (MI=0.123), EG, NL, KT, GN, AN, (MI=0.113, 0.111, 0.109, 0.105, and 0.103, respectively) and repulsive interactions (MI=0.100). Despite the lower MI scores, these features show a significant association with protein families and GO categories (Wilcoxon test p≤2.2x10 -16 ). Clustering of selected features reveals distinct patterns and relationships among protein families Hierarchical clustering grouped the dataset into 19 distinct clusters across eight families ( Fig. 4A ), revealing complex internal organization. The Short-chain dehydrogenases/reductases (SDR) family exhibited the most extensive dispersion, spanning seven clusters (C1, C6, C10, C11, C12, and C13), highlighting its functional diversity. Three distinct clusters (C17, C19, C20) encompassed the Cytochrome P450 family. Similarly, the Peptidase S1 family was distributed across three clusters (C8, C9, C16), suggesting potential functional specialization within this protease group. The Enoyl-CoA hydratase/isomerase family was found in two clusters (C2, C3), indicating possible sub-functionalization. In contrast, several protein families demonstrated a more focused distribution, each confined to a single cluster: Bacterial solute-binding protein 2 (C4), FPP/GGPP synthase (C14), Glycosyl hydrolase 5 (cellulase A) (C15), and Class-I aminoacyl-tRNA synthetase (C18). Download figure Open in new tab Fig. 4. Clustering and Visualization of Protein Families. A) Hierarchical clustering of protein families, depicting relationships and groupings based on structural and functional similarities. B) t-SNE plot depicting inter-family relationships within a reduced-dimensional space. C) Optimized projection plot highlighting the most significant features for each group, illustrating key attributes that differentiate protein families. Complementary t-SNE clustering analysis identified 9 groups ( Fig. 4B ), not only corroborating these groupings but also unveiling finer details of inter-group relationships. While hierarchical clustering effectively segmented proteins with similar functions, t-SNE provided a more nuanced separation. Occasional cross-family similarities were highlighted with some datapoints from different families appearing in unexpected clusters. Multivariate visualization further supported the role of interatomic interaction features ( Fig. 4C ), which were predominantly observed in the Cytochrome P450 and Glycosyl Hydrolase 5 (cellulase A) families (Wilcoxon p≤4.9x10 -4 , Supplementary Table 6 ). In contrast, dipeptide features were more prevalent among other protein families, indicating a varied functional landscape shaped by distinct sequence composition patterns. (Wilcoxon p≤0.031, See Supplementary Table 6 ). InteracTor’s feature selection effectively reduces dimensionality of models without compromising model performance Across all feature sets described in Table 2 , ensemble models, especially Histogram Gradient Boosting and Random Forest, consistently ranked among the top performers in protein family classification, achieving 0.71-0.80 F1-score on the test dataset after 80%-20% train-test split, indicating that the complex relationships captured by InteracTor’s features are effectively leveraged by these models’ ability to combine multiple decision trees ( Fig. 5A ). This trend was consistent across different accuracy metrics such as accuracy, precision, recall, and MCC (Supplementary Tables 7A and 7B) . CPAASC was the feature type that showed the lowest accuracy across all feature subsets (Tukey’s post hoc test p≤0.05, Supplementary Table 7A, 8 and Fig.5B ). MI score-based feature selection and further F1 score comparison across models showed that reducing InteracTor’s feature set to the top 100-500 features achieved comparable performance to using all 18,296 features (Tukey’s post hoc test p≥0.82). The smallest feature set selected via MI score (top 100) showed similar performance relative to larger subsets as well (Tukey’s post hoc test p≥0.12 versus top 200-500, Supplementary Table 8 ). These results highlight the effectiveness of our feature selection approach for optimizing model performance and interpretability. Download figure Open in new tab Fig. 5. Machine learning benchmark across different feature sets. A) F1 scores were computed for several classification models (shown as colored bars) across different feature sets. Feature sets were ranked by their impact on model performance, with higher F1 scores indicating superior predictive accuracy. B) Performance of the best model (Histogram Gradient Boosting) across different feature sets, showing mean validation accuracy after 5 runs. XAI reveals key features for protein family prediction models We employed eXplainable Artificial Intelligence (XAI) techniques, specifically the SHapley Additive exPlanations (SHAP) method( 20 ), to analyze the impact of InteracTor’s features on machine learning models for protein family prediction. The top 20 features showed SHAP values ranging from 0.25 to 1.75 and showed heterogenous contributions to the models’ predictions across different protein families ( Fig. 6 , Supplementary Fig. S2 ). The most important features included interatomic interactions and physicochemical properties, such as surface tension and hydrophobicity, as well as specific atomic interactions like repulsive interactions and hydrogen bonds. Additionally, certain peptide composition patterns, such as those involving amino acids P, CLG, PP, H, and W, were also among the most important features. These findings indicate that a combination of multimodal feature types, including structural properties, interatomic interactions, and amino acid composition, critically affects the model’s capacity to differentiate protein families. Download figure Open in new tab Fig. 6. Feature Importance (SHAP) in Protein Family Classification. A-G) The plot ranks features by their impact on model predictions, illustrating contribution distributions. Each dot represents a protein, with colors indicating feature values: red for high, blue for low. The horizontal spread of dots reflects the density of examples at given feature values. Negative SHAP values (left) indicate a reduced likelihood of classification, while positive SHAP values (right) suggest an increased likelihood. SHAP values were computed using family-specific explicands and baselines, ensuring consistent cross-group comparisons. h) Classification model performance measured by the F1 Score for each protein family. The bars represent the mean F1 Score obtained from 10 independent runs, with error bars corresponding to the standard deviation. The features were ranked differently by SHAP values across different protein families, with each protein family type exhibiting a unique association pattern. For instance, London dispersion forces, hydrogen bonds, and internal hydrophobicity had a positive impact on predictions for the Cytochrome P450 family ( Fig. 6A , Supplementary Fig. S2A ). However, these same features showed negative impacts for other families, including Bacterial solute-binding protein 2, Class-I aminoacyl-tRNA synthetases, and Short-chain dehydrogenases reductases ( Fig. 6B-G , Supplementary Fig. S2B-G ). FPP/GGPP synthases also exhibited a distinct pattern and ranking of SHAP values relative to other families, with key contributors from dipeptide DD, internal hydrophobicity, london dispersion forces, and several specific amino acid combinations including tripeptide RRG and monopeptide H ( Fig. 6B , Supplementary Fig. S2B ). DISCUSSION A novel algorithm for improved characterization of protein structural properties Our protein feature extraction algorithm distinguishes itself by focusing on the tertiary structure of proteins, unlike other approaches that rely on primary and secondary structures( 10 , 32 , 33 ). By analyzing three-dimensional relationships, the algorithm captures intramolecular interactions that underpin protein structure and function. Extracting features directly from the tertiary structure allows the algorithm to uncover additional patterns in the chemical properties of amino acids and their interactions, offering a comprehensive understanding of the structural and functional dynamics of proteins( 34 ). This enhanced perspective provides new insights into the complex mechanisms governing protein behavior, facilitating advancements in protein engineering and drug design( 35 ). The graphical representations ( Fig. 3 - 6 ) provide a comprehensive overview of protein characteristics as well, assisting in the visualization and interpretation of complex structural and physicochemical properties across protein families. The feature selection process, guided by MI scores, was crucial in identifying the most effective features for distinguishing protein families ( Fig. 3 ). Prominent peaks in MI scores indicated that features such as hydrogen bonds, total surface tension, and contact hydrophobicity are particularly influential in differentiating protein families. This prioritization of key features enhances our ability to effectively distinguish between various protein structural and functional groups. The identification of 9 out of 11 interatomic interactions features among the top 12 features with the highest MI scores underscores their importance in differentiating protein families. This finding defines the importance of specific molecular interactions, such as hydrogen bonds and hydrophobic contacts, in defining protein structure and function, providing valuable insights into the fundamental principles governing protein family diversity. Fig. 4 shows clear separations of specific clusters, highlighting the effectiveness of the approach in identifying grouping patterns among protein families. The groupings not only reflected complex interatomic relationships between different features but also demonstrated the formation of distinct clusters within dipeptide feature groups. The Peptidase S1 family, Glycosyl hydrolase 5 (cellulase A) family and Short-chain dehydrogenases/reductases (SDR) family demonstrated clustering of dipeptide feature groups. This result confirms the method’s ability to highlight relevant functional patterns and distinguish between different protein family categories based on the observed interatomic interactions. Clustering reveals complex structure and functional diversity in protein families Hierarchical clustering effectively delineated differences among protein families such as short-chain dehydrogenases/reductases (SDR), Cytochrome P450, Enoyl-CoA hydratase/isomerase, Bacterial solute-binding protein 2, Class-I aminoacyl-tRNA synthetase, Glycosyl hydrolase 5 (cellulase A), Peptidase S1, and FPP/GGPP synthase. Complementary analysis with t-SNE allowed the detection of subtle similarities and differences among the protein families and offering deeper insights into their functional connections ( Fig. 4 , Supplementary Table 6 ). Multivariate visualization allowed dynamic adjustments to identify unique aspects of protein families, including specific features contributing to the clustering process, yielding valuable insights into the functional diversity within and between protein families. This integrative approach led to a deeper understanding of the complex relationships and functional diversity inherent across protein families, revealing details on the interplay of intermolecular interactions, physicochemical properties, and composition of proteins. This holistic methodology not only enhances our comprehension of protein family dynamics but also establishes a robust framework for future research. The extracted features have the potential to directly impact the performance of machine learning and deep learning models by offering a new feature engineering and feature selection tool. By providing new insights on protein structural and functional properties, our approach supports advancements in structural biology, synthetic biology and in the characterization proteins of unknown function (PUFs)( 36 – 39 ). Additionally, these features can contribute to emerging fields such as quantum biology( 40 – 44 ), facilitating a deeper understanding of biological roles and interactions across diverse protein families. XAI reveals key features for classification of protein function families Our XAI results show that a combination of physicochemical properties, atomic interactions, and peptide motifs enhances protein family classification accuracy. This approach underscores the importance of integrating protein structural, physicochemical and sequence features in computational modeling, assembling groundwork for advanced protein analysis methods. By integrating these features, we open a new pathway for understanding protein structure, activity, and function. Feature selection and XAI analysis revealed that atomic-level interactions, specifically hydrogen bonds and London dispersion forces, in conjunction with distinct peptide composition patterns (e.g., proline, cysteine-leucine-glycine, and diproline sequences) were pivotal for identifying protein families. Moreover, the application of XAI methodologies elucidated the differential impact of these features across various protein families ( Fig. 6 ). The heterogeneity in feature importance can be attributed to the structural and functional diversity of the features among protein families. For instance, the importance of London dispersion forces( 45 ) and internal hydrophobicity( 46 ) in Cytochrome P450 classification emphasizes the crucial role these interactions play in maintaining its tertiary structure. Conversely, the inverse correlation of these same features in families such as Peptidase S1 indicates that alternative factors such as hydrophobicity, the mono-dipeptide, and the notable presence of specific catalytic domains may exert a more substantial influence on classification outcomes( 47 ) ( Supplementary Fig. S2F ). Our findings align with previous studies that emphasize the significance of atomic interactions and physicochemical properties in protein classification( 5 , 10 – 12 , 17 – 19 ). Other similar features based in sequence property have been widely reviewed( 48 , 49 ). Our proposed set of interatomic interaction features were often among the top ranked features contributing to the accurate classification of protein families, complementing other feature types ( Supplementary Tables 5-8 ). By leveraging XAI, we successfully quantified and interpreted the specific impact of each feature- an aspect that conventional approaches could not comprehensively address-offering deeper insights into the biological mechanisms underlying the model’s performance. The capacity to identify key features for protein classification has implications for drug design and protein engineering because understanding molecular interactions is decisive. InteracTor enables interpretable modeling tasks such as protein feature engineering, feature selection, and XAI, supporting applications like drug discovery (e.g., binding site analysis), protein function prediction, and structure-guided protein design. InteracTor applies XAI principles—such as feature importance attribution and structural interpretability—to map interatomic interactions to functional outcomes. This bridges the gap between black-box protein models and actionable insights, enabling targeted protein engineering (e.g., optimizing binding affinity or stability by prioritizing CPAASC features) and error diagnosis (e.g., flagging repulsive VDW clashes in misfolded designs). Mechanistic insights on protein structure, activity, and function The features extracted by InteracTor provide critical insights into the structural and functional diversity of analyzed protein families. Hydrogen bonds were among the top predictors for classifying SDR proteins. SDR proteins have a conserved α/β folding pattern with a central beta sheet flanked by 2-3 α-helices from each side, forming a classical Rossmann-fold motif for nucleotide binding. Hydrogen bonds are known to be essential in the stabilization of the Rossmann fold and the NAD(P)(H)-binding region of SDRs, which is consistent with our findings( 50 ),( 51 ). ASA and total hydrophobic contacts were key features in the classification of Cytochrome P450 family. The importance of hydrophobic contacts in Cytochrome P450 function has been highlighted by studies showing that hydrophobic residues are pivotal in complex formation with their redox partners ( 52 ). Cytochrome P450s also have deeply buried active sites that are connected to the solvent by a network of channels exiting at the distal surface of the protein, which could be reflected by the high contribution of ASA in the classification of this protein family( 53 ). The Bacterial solute-binding protein 2 family displayed a balance of repulsive interactions and internal tension, likely linked to their dynamic conformational states required for substrate transport( 54 ). Proteins from the Class-I aminoacyl-tRNA synthetase family were characterized by high deformation effect, supporting their complex multi-step role in amino acid activation and tRNA binding( 55 ). The Glycosyl hydrolase 5 (cellulase A) family exhibited pronounced hydrophobic contacts and surface tension, aligning with their function in breaking down robust carbohydrate polymers( 56 ). Hydrophobicity-related features were key predictors for the Peptidase S1 family. Chymotrypsin-like enzymes within the S1 family have a hydrophobic S1 pocket, which allows them to cleave peptide bonds following medium to large hydrophobic amino acids such as tyrosine, phenylalanine, and tryptophan, which is consistent with our findings( 57 ). The Peptidase S1 family also contains a catalytic site that is typically preceded by a block of hydrophobic residues as well, which is also consistent with our results( 58 ). Lastly, the FPP/GGPP synthase family demonstrated a notable deformation effect, which is consistent with their known versatility by accommodating diverse substrates during isoprenoid biosynthesis( 59 ) These observations highlight InteracTor’s utility in characterizing protein families based on their global structural features, offering a foundation for future studies exploring interaction-specific mechanisms within these families. RESOURCE AVAILABILITY Data and code availability InteracTor is available as an open-source Python package on GitHub ( https://github.com/Dias-Lab/InteracTor ), providing a wealth of details in both code and documentation. The repository includes comprehensive implementation details, example datasets, and usage guidelines to ensure transparency and reproducibility. Users can explore, modify, and contribute to enhance its capabilities. AUTHOR CONTRIBUTIONS R.D. designed and supervised the study and contributed to manuscript writing. M.F.R.R and M.K. contributed to manuscript writing. J.C.F.S. developed the algorithm, built the computational tools, performed the analysis, and wrote the manuscript. L.S and N.S. were responsible for the data processing and contributed to manuscript writing. M.E. and R.H. contributed to software testing and manuscript writing. All authors read and approved the final manuscript. DECLARATION OF INTERESTS The authors declare no conflicts of interest. METHODS Algorithm overview InteracTor algorithm consists of a sequence of steps for analyzing protein structures and calculating various features from proteins’ interatomic interactions, physicochemical properties of residues, and peptide composition ( Fig. 1 ). The following is a description of each step: Step 1 . Extract atom, residue, and sequence information from PDB file ( Fig. 1A ): this process involves parsing the Protein Data Bank (PDB) file to obtain the atomic types, 3D coordinates, and the amino acid sequence of the protein. Step 2 . Extract additional atomic information from MOL2 file ( Fig. 1A ), including covalent bond mapping and types (e.g., single, double, triple bonds, aromatic rings, etc.) and sp 2 hybridization. Step 3. Calculate atom-atom distances ( Fig. 1B ): the algorithm computes van der Waals (VDW) radii and distances between pairs of atoms not covalently bound within the protein structure, which is used to identify potential interatomic interactions within the protein structure. A distance threshold, as described by Dias et al.,( 5 ) is applied to determine whether atoms are interacting. Step 4 . Calculate geometric centers ( Fig. 1B ): additional properties are extracted from atoms selected as potentially participating in non-covalent interactions (step 3) and covalently bound to multiple atoms (step 2). This includes the calculation of geometric centers between the selected atoms and their respective covalently bound atoms. These centers are utilized to calculate angles between hydrogen bond donor and acceptor atoms (step 5). Step 5. Calculate angles ( Fig. 1B ): the algorithm then computes the angles between atoms or groups of atoms in the protein structure, which is used to evaluate hydrogen bonding geometry. Step 6 . Compute interatomic interaction features and protein physicochemical properties described in Table 1 and Fig. 1C . Step 7. Extract compositional features ( Supplementary Table 2 and Fig. 1C ): InteracTor calculates the frequencies of single amino acids (monopeptide), pairs of amino acids (dipeptide), and triplets of amino acids (tripeptide) in the protein sequence. Step 8 . Extract CPAASC frequencies ( Supplementary Table 1 and Fig. 1C ): the algorithm computes the frequencies of several physicochemical properties associated with the side chains of the amino acids in the protein sequence. Step 9. Write results and postprocessing ( Fig. 1D ): the final step is to write the features computed in steps 6, 7 and 8 to an output file for further analysis or use in other applications. The toolkit also includes example scripts for downstream analyses such as feature selection via MI scoring, dimensionality reduction via PCA, t-SNE, UMAP, hierarchical clustering and visualization using heatmaps, and protein function classification using machine learning and XAI. Our toolkit extracts and encodes protein structural features based on physico-chemical properties, amino acids composition, and interatomic interactions. By building upon our previous methods for predicting protein-ligand binding affinity( 5 ), we modified the algorithm to analyze residue-residue interactions within protein structures ( Table 1 ). We also included the calculation of CPAASC frequencies( 17 ), ( Supplementary Table 1 ), monopeptide( 18 ), dipeptide, and tripeptide composition( 19 ) ( Supplementary Table 2 ). In addition to the 20 classic amino acids, our algorithm supports rare residues found in distinct biological systems, including selenocysteine, pyrrolysine, and N-methylvaline. These amino acids, although rare, are key in specific biological processes( 60 ),( 61 ). Structural datasets and preprocessing Dataset for clustering analyses. We used 20,877 protein structures from the PDB-REDO database( 62 ), having eliminated redundant (50% sequence similarity) and small proteins (<50 residues) using Biopython and in-house scripts (see Data and Code Availability). PDB-REDO was used to enhance structural data quality, improving machine learning model reliability by minimizing errors and inconsistencies (e.g., low resolution, missing atoms, etc.). This approach aligns with research showing that prioritizing data quality over quantity leads to better prediction performance and model robustness( 62 ). Including distant protein relatives with lower identity cutoffs in the study can provide valuable insights into functional conservation and evolutionary relationships as proteins with low sequence identity can still share similar tertiary structures and functions( 62 ). PDB files were converted to MOL2 format using Open Babel( 63 ). We then applied the InteracTor algorithm to extract features from both PDB and MOL2 files. Datasets for machine learning and XAI. We further filtered out proteins with more than 50% sequence identity to minimize data leakage in our machine learning and XAI experiments. We also filtered out structures with low resolution (>2.5 Å) and less than 100 residues in order to reduce noise and potential outliers. Datasets selected for demonstration. We utilized UniProt’s application programming interface (API)( 64 ) to extract Gene Ontology( 65 ) (GO) terms and protein family names based on the PDB accession number( 66 ) to facilitate analysis and demonstration of use cases. We performed power analysis to determine the minimum sample size needed per protein family and GO term for further demonstration of use cases ( Table 1 , 2 ). For the demonstration of use cases, we selected protein family categories with at least 30 representatives in our dataset to ensure sufficient statistical power for downstream analyses. Similarly, we applied a minimum threshold of 90 annotations for Gene Ontology (GO) terms but further restricted selection to terms directly associated with the binding mechanisms of proteins (e.g., ligand type) to further assess the potential of InteracTor for profiling of protein functions directly associated with ligand or substrate binding. By selecting protein families and GO terms with enough sample sizes, we ensure that our use cases effectively demonstrate our approach’s capabilities. Mapping and power analysis scripts are available in GitHub as well (see Data and Code Availability). Feature selection and clustering analysis We applied MI scoring to quantify the relevance of each feature in distinguishing between different protein families and GO categories( 67 ). We ranked the features by their respective MI scores and selected the top 100 most informative features for further analyses. We applied PCA to reduce the dimensionality of the data generated by our toolkit and evaluated how the primary components capture the variability among protein families and GO terms( 68 ). In addition to PCA, we also applied t-SNE( 69 , 70 ) and Freeviz( 71 ) to reduce dimensionality while preserving nonlinear relationships and local structures in data. We used violin plots to visualize the distribution and density of the extracted features across different protein families and GO terms, providing insights into the distribution and central tendencies of the data. We performed hierarchical clustering using Pearson correlation as the distance metric and the complete linkage method for cluster formation( 70 ). This approach allowed us to identify and visualize the hierarchical relationships between protein families based on similarities among the features extracted by our toolkit. The dimensionality reduction and clustering results were visualized with FreeViz. Machine learning benchmark We generated 11 feature sets to assess their contribution to improving machine learning performance in predicting structural properties. Five datasets comprised individual feature types: interatomic interactions, CPAASC, monopeptides, dipeptides, and tripeptides ( Table 2 ). Another five datasets were created using the top 100, 200, 300, 400, and 500 features ranked by MI scores, incorporating a mixture of the five feature types ( Supplementary Table 5 ). The eleventh dataset integrated all features extracted by InteracTor ( Table 2 ). This approach allowed us to evaluate the relative importance of different feature combinations in enhancing predictive accuracy. We tested Machine Learning (ML) models to assess the performance of InteracTor’s features in the prediction of protein families. In this experiment, we used 43 ML algorithms implemented in the Python library Scikit-learn ( Supplementary Table 7D )( 72 ). To evaluate model performance, we randomly shuffled the data, performed 80%-20% train-test split and measured multiclass classification accuracy, precision, F1 score, Mathew’s correlation coefficient (MCC), and precision for each combination of feature set and algorithm. This process was repeated 5 times. We also performed One-Vs-Rest-Classifier cross validation on the best model and feature set to compute F1 scores for each protein family class( 70 ). eXplainable AI (XAI) methods We calculated SHAP (SHapley Additive exPlanations)( 73 ) values to quantify the individual feature contributions in predicting protein families using best-performing model (HistGradientBoostingClassifier). SHAP values were extracted from the test set using the trained model. We then generated SHAP summary plots using the summary plot function from the SHAP library( 20 ). To create SHAP bar plots, we calculated the total absolute SHAP values for each feature and determined the effect direction by computing the Pearson correlation between SHAP values and the model’s predicted probabilities Statistical analysis Statistical Power tests( 74 ) were performed using the TTestIndPower from the statsmodels package in Python to compute the Cohen’s d effect sizes for comparisons between groups (effect size = 0.8, alpha error = 0.05, power = 0.8). The Wilcoxon( 75 ) test was used to compare the means and medians between feature sets and models. We ran Tukey’s test( 76 ) pairwise comparisons among F1 scores to identify feature sets that significantly contribute to model performance. We applied p≤0.05 as statistical significance threshold across all statistical tests. SUPPLEMENTARY FIGURE LEGENDS Fig. S1. Principal component analysis of protein families and GO terms. A) PCA plot illustrating the variance captured by the first three principal components across different protein families. B) PCA plot depicting the variance explained by the first three principal components across Gene Ontology (GO) terms. Fig. S2. Global feature importance in Protein Family Classification. Mean absolute SHAP values were computed for each protein family, representing the overall impact of features on each protein family. The direction of the impact was computed based on the correlation between the SHAP values and the likelihood of the predicted class: red for positive correlation, blue for negative correlation. Subplots (A-G) correspond to distinct protein families. SUPPLEMENTARY TABLE LEGENDS Supplementary Table 1 : Protein physicochemical features based on chemical properties of amino acid side chains (CPAASC). Supplementary Table 2 : Protein sequence composition features. Supplementary Table 3 : Distribution of selected protein families. Supplementary Table 4: Distribution of selected GO terms. Supplemental Table 5. MI Scores for protein families and GO terms. Supplemental Table 6. Wilcoxon p-values for pairwise comparisons between protein families. Supplemental Table 7. a) Feature Set Performance: A comprehensive ranking of all evaluated feature sets across all tested algorithms, sorted by F1-score. b) Algorithm Performance Overview: A ranking of algorithms based on their mean F1-score across all feature sets, offering insights into overall algorithm efficacy. c) Top-Performing Algorithms by Feature Set: For each feature set, the five best-performing algorithms are highlighted, facilitating the identification of optimal algorithm-feature set combinations. d) Algorithm Characteristics and Performance: A detailed ranking of algorithms, including descriptions of their underlying principles, classification strategies, and performance metrics. Supplementary Table 8: Tukey’s Test for Feature Significance Based on F1-Score. This table presents the results of Tukey’s post-hoc test, assessing differences and statistical significance among features based on their F1-scores. ACKNOWLEDGEMENTS The authors gratefully acknowledge UF Research Computing for providing computational resources and support that have contributed to the research reported in this publication ( http://www.rc.ufl.edu ). This work was also supported by UF LIFT AI funding of University of Florida’s Institute of Food and Agricultural Sciences (UF/IFAS). REFERENCES 1. ↵ Varadi M , Anyango S , Deshpande M , Nair S , Natassia C , Yordanova G , et al. AlphaFold Protein Structure Database: massively expanding the structural coverage of protein-sequence space with high-accuracy models . Nucleic Acids Research . 2022 Jan 7; 50 ( D1 ): D439 – 44 . OpenUrl CrossRef PubMed 2. ↵ Baek M , DiMaio F , Anishchenko I , Dauparas J , Ovchinnikov S , Lee GR , et al. Accurate prediction of protein structures and interactions using a three-track neural network . Science . 2021 Aug 20; 373 (6557):871–6. 3. ↵ Wang X , Li F , Xu J , Rong J , Webb GI , Ge Z , et al. ASPIRER: a new computational approach for identifying non-classical secreted proteins based on deep learning . Briefings in Bioinformatics . 2022 Mar 10; 23 ( 2 ):bbac031. 4. ↵ Zhang Y , Yu S , Xie R , Li J , Leier A , Marquez-Lago TT , et al. PeNGaRoo, a combined gradient boosting and ensemble learning framework for predicting non-classical secreted proteins . Hancock J , editor. Bioinformatics . 2020 Feb 1; 36 ( 3 ): 704 – 12 . OpenUrl CrossRef PubMed 5. ↵ Dias R , Kolazckowski B . Different combinations of atomic interactions predict protein-small molecule and protein-DNA/RNA affinities with similar accuracy . Proteins . 2015 Nov ; 83 ( 11 ): 2100 – 14 . OpenUrl CrossRef PubMed 6. ↵ Hong X , Tong X , Xie J , Liu P , Liu X , Song Q , et al. An updated dataset and a structure-based prediction model for protein– RNA binding affinity . Proteins . 2023 Sep ; 91 ( 9 ): 1245 – 53 . OpenUrl CrossRef PubMed 7. ↵ Zhang H , Saravanan KM , Zhang JZH . DeepBindGCN: Integrating Molecular Vector Representation with Graph Convolutional Neural Networks for Protein–Ligand Interaction Prediction . Molecules . 2023 Jun 10; 28 ( 12 ): 4691 . OpenUrl CrossRef PubMed 8. ↵ Torng W , Altman RB . 3D deep convolutional neural networks for amino acid environment similarity analysis . BMC Bioinformatics . 2017 Dec ; 18 ( 1 ): 302 . OpenUrl CrossRef PubMed 9. ↵ Shroff R , Cole AW , Diaz DJ , Morrow BR , Donnell I , Annapareddy A , et al. Discovery of Novel Gain-of-Function Mutations Guided by Structure-Based Deep Learning . ACS Synth Biol . 2020 Nov 20; 9 ( 11 ): 2927 – 35 . OpenUrl CrossRef PubMed 10. ↵ Chen Z , Liu X , Zhao P , Li C , Wang Y , Li F , et al. iFeatureOmega: an integrative platform for engineering, visualization and analysis of features from molecular sequences, structural and ligand data sets . Nucleic Acids Research . 2022 Jul 5; 50 ( W1 ): W434 – 47 . OpenUrl CrossRef PubMed 11. ↵ Pande A , Patiyal S , Lathwal A , Arora C , Kaur D , Dhall A , et al. Pfeature: A Tool for Computing Wide Range of Protein Features and Building Prediction Models . Journal of Computational Biology . 2023 Feb 1; 30 ( 2 ): 204 – 22 . OpenUrl CrossRef PubMed 12. ↵ Guevara-Barrientos D , Kaundal R . ProFeatX: A parallelized protein feature extraction suite for machine learning . Computational and Structural Biotechnology Journal . 2023 ; 21 : 796 – 801 . OpenUrl CrossRef 13. ↵ Hassija V , Chamola V , Mahapatra A , Singal A , Goel D , Huang K , et al. Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence . Cogn Comput . 2024 Jan ; 16 ( 1 ): 45 – 74 . OpenUrl CrossRef 14. ↵ Gao J , Bosco DA , Powers ET , Kelly JW . Localized thermodynamic coupling between hydrogen bonding and microenvironment polarity substantially stabilizes proteins . Nat Struct Mol Biol . 2009 Jul ; 16 ( 7 ): 684 – 90 . OpenUrl CrossRef PubMed 15. ↵ Coulocheri SA , Pigis DG , Papavassiliou KA , Papavassiliou AG . Hydrogen bonds in protein– DNA complexes: Where geometry meets plasticity . Biochimie . 2007 Nov ; 89 ( 11 ): 1291 – 303 . OpenUrl CrossRef PubMed Web of Science 16. ↵ Biomedical Informatics (P) Ltd , Irulan Sandy Annex, Puducherry 607 402, India, Nilofer C, School of Bio Sciences and Technology, VIT University, Vellore, Tamil Nadu, India, Sukhwal A, National Centre for Biological Sciences, TIFR, UAS-GKVK Campus, Bangalore, Karnataka, India, SASTRA University, Thanjavur, Tamil Nadu, India, et al. Protein-protein interfaces are vdW dominant with selective H-bonds and (or) electrostatics towards broad functional specificity . Bioinformation . 2017 Jun 30; 13 ( 06 ): 164 – 73 . OpenUrl CrossRef PubMed 17. ↵ Carvalho TFM , Silva JCF , Calil IP , Fontes EPB , Cerqueira FR. Rama: a machine learning approach for ribosomal protein prediction in plants. Sci Rep . 2017 Nov 24;7(1):16273 . 18. ↵ Silva JCF , Ferreira MA , Carvalho TFM , Silva FF , De A. Silveira S , Brommonschenkel SH , et al. RLPredictiOme, a Machine Learning-Derived Method for High-Throughput Prediction of Plant Receptor-like Proteins , Reveals Novel Classes of Transmembrane Receptors. IJMS . 2022 Oct 12; 23 ( 20 ): 12176 . OpenUrl PubMed 19. ↵ Silva JCF , Carvalho TFM , Fontes EPB , Cerqueira FR . Fangorn Forest (F2): a machine learning approach to classify genes and genera in the family Geminiviridae . BMC Bioinformatics . 2017 Dec ; 18 ( 1 ): 431 . OpenUrl CrossRef PubMed 20. ↵ Lundberg SM , Erion G , Chen H , DeGrave A , Prutkin JM , Nair B , et al. From local explanations to global understanding with explainable AI for trees . Nat Mach Intell . 2020 Jan 17; 2 ( 1 ): 56 – 67 . OpenUrl CrossRef PubMed 21. Pace CN , Fu H , Fryar KL , Landua J , Trevino SR , Shirley BA , et al. Contribution of Hydrophobic Interactions to Protein Stability . Journal of Molecular Biology . 2011 May ; 408 ( 3 ): 514 – 28 . OpenUrl CrossRef PubMed 22. Sung S . Peptide folding driven by V an der W aals interactions . Protein Science . 2015 Sep ; 24 ( 9 ): 1383 – 8 . OpenUrl CrossRef PubMed 23. Bao G . Protein Mechanics: A New Frontier in Biomechanics . Exp Mech . 2009 Feb ; 49 ( 1 ): 153 – 64 . OpenUrl CrossRef PubMed 24. Herschlag D , Pinney MM . Hydrogen Bonds: Simple after All? Biochemistry . 2018 Jun 19; 57 ( 24 ): 3338 – 52 . OpenUrl CrossRef PubMed 25. Sung S . Peptide folding driven by V an der W aals interactions . Protein Science . 2015 Sep ; 24 ( 9 ): 1383 – 8 . OpenUrl CrossRef PubMed 26. Walters ET , Mohebifar M , Johnson ER , Rowley CN . Evaluating the London Dispersion Coefficients of Protein Force Fields Using the Exchange-Hole Dipole Moment Model . J Phys Chem B . 2018 Jul 5; 122 ( 26 ): 6690 – 701 . OpenUrl 27. Van Gils JHM , Gogishvili D , Van Eck J , Bouwmeester R , Van Dijk E , Abeln S. How sticky are our proteins? Quantifying hydrophobicity of the human proteome . Gromiha M , editor. Bioinformatics Advances . 2022 Jan 10; 2 ( 1 ): vbac002 . OpenUrl 28. Wei AP , Herron JN , Andrade JD. The Role of Protein Structure in Surface Tension Kinetics . In: Crommelin DJA , Schellekens H , editors. From Clone to Clinic [Internet] . Dordrecht: Springer Netherlands ; 1990 [cited 2025 Feb 15]. p. 305 – 13 . (Developments in Biotherapy; vol. 1). Available from: http://link.springer.com/10.1007/978-94-011-3780-5_38 29. Sashi P , Ramakrishna D , Bhuyan AK . Dispersion Forces and the Molecular Origin of Internal Friction in Protein . Biochemistry . 2016 Aug 23; 55 ( 33 ): 4595 – 602 . OpenUrl CrossRef PubMed 30. Shrake A , Rupley JA . Environment and exposure to solvent of protein atoms. Lysozyme and insulin . Journal of Molecular Biology . 1973 Sep ; 79 ( 2 ): 351 – 71 . OpenUrl CrossRef PubMed Web of Science 31. ↵ Gaudet P , Dessimoz C. Gene Ontology: Pitfalls, Biases, and Remedies . In: Dessimoz C , Škunca N , editors. The Gene Ontology Handbook [Internet] . New York, NY: Springer New York ; 2017 [cited 2025 Feb 15]. p. 189–205. (Methods in Molecular Biology; vol. 1446). Available from: http://link.springer.com/10.1007/978-1-4939-3743-1_14 32. ↵ Cao J , Xiong L . Protein Sequence Classification with Improved Extreme Learning Machine Algorithms . BioMed Research International . 2014 ;2014: 1 – 12 . 33. ↵ Markin CJ , Mokhtari DA , Sunden F , Appel MJ , Akiva E , Longwell SA , et al. Revealing enzyme functional architecture via high-throughput microfluidic enzyme kinetics . Science . 2021 Jul 23; 373 ( 6553 ): eabf8761. OpenUrl Abstract / FREE Full Text 34. ↵ Holehouse AS , Kragelund BB . The molecular basis for cellular function of intrinsically disordered protein regions . Nat Rev Mol Cell Biol . 2024 Mar ; 25 ( 3 ): 187 – 211 . OpenUrl CrossRef PubMed 35. ↵ De Juan D , Pazos F , Valencia A . Emerging methods in protein co-evolution . Nat Rev Genet . 2013 Apr ; 14 ( 4 ): 249 – 61 . OpenUrl CrossRef PubMed 36. ↵ Poudel S , Cope AL , O’Dell KB , Guss AM , Seo H , Trinh CT , et al. Identification and characterization of proteins of unknown function (PUFs) in Clostridium thermocellum DSM 1313 strains as potential genetic engineering targets . Biotechnol Biofuels . 2021 Dec ; 14 ( 1 ): 116 . OpenUrl CrossRef PubMed 37. Park J , Lappe M , Teichmann SA . Mapping protein family interactions: intramolecular and intermolecular protein family interaction repertoires in the PDB and yeast11Edited by J. Karn . Journal of Molecular Biology . 2001 Mar ; 307 ( 3 ): 929 – 38 . OpenUrl CrossRef PubMed Web of Science 38. Hoffmann SA , Cai Y . Engineering stringent genetic biocontainment of yeast with a protein stability switch . Nat Commun . 2024 Feb 5; 15 ( 1 ): 1060 . OpenUrl CrossRef PubMed 39. ↵ Nadzirin N , Firdaus-Raih M . Proteins of Unknown Function in the Protein Data Bank (PDB): An Inventory of True Uncharacterized Proteins and Computational Tools for Their Analysis . IJMS . 2012 Oct 8; 13 ( 10 ): 12761 – 72 . OpenUrl CrossRef PubMed 40. ↵ Pal S , Bhattacharya M , Lee SS , Chakraborty C . Quantum Computing in the Next-Generation Computational Biology Landscape: From Protein Folding to Molecular Dynamics . Mol Biotechnol . 2024 Feb ; 66 ( 2 ): 163 – 78 . OpenUrl CrossRef PubMed 41. Ha P , Kwak JH , Zhang Y , Shi J , Tran L , Liu TP , et al. Bisphosphonate conjugation enhances the bone-specificity of NELL-1-based systemic therapy for spaceflight-induced bone loss in mice . npj Microgravity . 2023 Sep 18; 9 ( 1 ): 75 . OpenUrl CrossRef PubMed 42. Doga H , Raubenolt B , Cumbo F , Joshi J , DiFilippo FP , Qin J , et al. A Perspective on Protein Structure Prediction Using Quantum Computers . J Chem Theory Comput . 2024 May 14; 20 ( 9 ): 3359 – 78 . OpenUrl CrossRef PubMed 43. Menezes AA , Cumbers J , Hogan JA , Arkin AP . Towards synthetic biological approaches to resource utilization on space missions . J R Soc Interface . 2015 Jan ; 12 ( 102 ): 20140715 . OpenUrl CrossRef PubMed 44. ↵ Mokhtari M , Khoshbakht S , Ziyaei K , Akbari ME , Moravveji SS . New classifications for quantum bioinformatics: Q-bioinformatics, QCt-bioinformatics, QCg-bioinformatics, and QCr-bioinformatics . Briefings in Bioinformatics . 2024 Jan 22; 25 ( 2 ):bbae074. 45. ↵ Reimers JR , Ford MJ , Marcuccio SM , Ulstrup J , Hush NS . Competition of van der Waals and chemical forces on gold–sulfur surfaces and nanoparticles . Nat Rev Chem . 2017 Feb 8; 1 ( 2 ): 0017 . OpenUrl CrossRef 46. ↵ Shah MB , Wilderman PR , Liu J , Jang HH , Zhang Q , Stout CD , et al. Structural and Biophysical Characterization of Human Cytochromes P450 2B6 and 2A6 Bound to Volatile Hydrocarbons: Analysis and Comparison . Molecular Pharmacology . 2015 Apr ; 87 ( 4 ): 649 – 59 . OpenUrl Abstract / FREE Full Text 47. ↵ Rigolet P , Mechin I , Delage MM , Chich JF . The Structural Basis for Catalysis and Specificity of the X-Prolyl Dipeptidyl Aminopeptidase from Lactococcus lactis . Structure . 2002 Oct ; 10 ( 10 ): 1383 – 94 . OpenUrl CrossRef PubMed 48. ↵ Greener JG , Kandathil SM , Moffat L , Jones DT . A guide to machine learning for biologists . Nat Rev Mol Cell Biol . 2022 Jan ; 23 ( 1 ): 40 – 55 . OpenUrl CrossRef PubMed 49. ↵ Silva JCF , Teixeira RM , Silva FF , Brommonschenkel SH , Fontes EPB . Machine learning approaches and their current application in plant molecular biology: A systematic review . Plant Science . 2019 Jul ; 284 : 37 – 47 . OpenUrl CrossRef PubMed 50. ↵ Demange P , Joly E , Marcoux J , Zanon PR , Listunov D , Rullière P , et al. SDR enzymes oxidize specific lipidic alkynylcarbinols into cytotoxic protein-reactive species . eLife . 2022 May 10; 11 : e73913 . OpenUrl CrossRef PubMed 51. ↵ Kavanagh KL , Jörnvall H , Persson B , Oppermann U . Medium- and short-chain dehydrogenase/reductase gene and protein families: The SDR superfamily: functional and structural diversity within a family of metabolic and regulatory enzymes . Cell Mol Life Sci . 2008 Dec ; 65 ( 24 ): 3895 . OpenUrl CrossRef PubMed Web of Science 52. ↵ Kenaan C , Zhang H , Shea EV , Hollenberg PF . Uncovering the Role of Hydrophobic Residues in Cytochrome P450−Cytochrome P450 Reductase Interactions . Biochemistry . 2011 May 17; 50 ( 19 ): 3957 – 67 . OpenUrl CrossRef PubMed Web of Science 53. ↵ Urban P , Lautier T , Pompon D , Truan G . Ligand Access Channels in Cytochrome P450 Enzymes: A Review . IJMS . 2018 May 30; 19 ( 6 ): 1617 . OpenUrl CrossRef PubMed 54. ↵ Gennis RB . Pores, Channels and Transporters . In: Biomembranes [Internet] . New York, NY: Springer New York ; 1989 [cited 2025 Feb 15]. p. 270 – 322 . (Cantor CR, editor. Springer Advanced Texts in Chemistry). Available from: http://link.springer.com/10.1007/978-1-4757-2065-5_8 55. ↵ Delagoutte B . tRNA aminoacylation by arginyl-tRNA synthetase: induced conformations during substrates binding . The EMBO Journal . 2000 Nov 1; 19 ( 21 ): 5599 – 610 . OpenUrl Abstract / FREE Full Text 56. ↵ Pozzo T , Pasten JL , Karlsson EN , Logan DT . Structural and Functional Analyses of β-Glucosidase 3B from Thermotoga neapolitana: A Thermostable Three-Domain Representative of Glycoside Hydrolase 3 . Journal of Molecular Biology . 2010 Apr ; 397 ( 3 ): 724 – 39 . OpenUrl CrossRef PubMed 57. ↵ Scheidig AJ , Hynes TR , Pelletier LA , Wells JA , Kossiakoff AA . Crystal structures of bovine chymotrypsin and trypsin complexed to the inhibitor domain of alzheimer’s amyloid β-protein precursor (APPI) and basic pancreatic trypsin inhibitor (BPTI): Engineering of inhibitors with altered specificities . Protein Science . 1997 Sep ; 6 ( 9 ): 1806 – 24 . OpenUrl CrossRef PubMed Web of Science 58. ↵ Mizusawa N , Harada N , Iwata T , Ohigashi I , Itakura M , Yoshimoto K . Identification of protease serine S1 family member 53 as a mitochondrial protein in murine islet beta cells . Islets . 2022 Dec 31; 14 ( 1 ): 1 – 13 . OpenUrl CrossRef PubMed 59. ↵ Song S , Jin R , Chen Y , He S , Li K , Tang Q , et al. The functional evolution of architecturally different plant geranyl diphosphate synthases from geranylgeranyl diphosphate synthase . The Plant Cell . 2023 May 29; 35 ( 6 ): 2293 – 315 . OpenUrl CrossRef PubMed 60. ↵ Brugère JF , Atkins JF , O’Toole PW , Borrel G. Pyrrolysine in archaea: a 22nd amino acid encoded through a genetic code expansion . Robinson NP , editor. Emerging Topics in Life Sciences . 2018 Dec 14 ;2(4): 607 – 18 . OpenUrl 61. ↵ Turanov AA , Lobanov AV , Fomenko DE , Morrison HG , Sogin ML , Klobutcher LA , et al. Genetic Code Supports Targeted Insertion of Two Amino Acids by One Codon . Science . 2009 Jan 9; 323 ( 5911 ): 259 – 61 . OpenUrl Abstract / FREE Full Text 62. ↵ Joosten RP , Long F , Murshudov GN , Perrakis A . The PDB_REDO server for macromolecular structure model optimization . IUCrJ . 2014 Jul 1; 1 ( 4 ): 213 – 20 . OpenUrl CrossRef PubMed 63. ↵ O’Boyle NM , Banck M , James CA , Morley C , Vandermeersch T , Hutchison GR . Open Babel: An open chemical toolbox . J Cheminform . 2011 Dec ; 3 ( 1 ): 33 . OpenUrl CrossRef PubMed 64. ↵ The UniProt Consortium , Bateman A , Martin MJ , Orchard S , Magrane M , Agivetova R , et al. UniProt: the universal protein knowledgebase in 2021 . Nucleic Acids Research . 2021 Jan 8; 49 ( D1 ): D480 – 9 . OpenUrl CrossRef PubMed 65. ↵ Ashburner M , Ball CA , Blake JA , Botstein D , Butler H , Cherry JM , et al. Gene Ontology: tool for the unification of biology . Nat Genet . 2000 May ; 25 ( 1 ): 25 – 9 . OpenUrl CrossRef PubMed Web of Science 66. ↵ Berman HM . The Protein Data Bank . Nucleic Acids Research . 2000 Jan 1; 28 ( 1 ): 235 – 42 . OpenUrl CrossRef PubMed Web of Science 67. ↵ Kraskov A , Stögbauer H , Grassberger P . Estimating mutual information . Phys Rev E . 2004 Jun 23; 69 ( 6 ): 066138 . OpenUrl CrossRef 68. ↵ Jolliffe IT , Cadima J . Principal component analysis: a review and recent developments . Phil Trans R Soc A . 2016 Apr 13; 374 (2065):20150202. 69. ↵ Belkina AC , Ciccolella CO , Anno R , Halpert R , Spidlen J , Snyder-Cappione JE . Automated optimized parameters for T-distributed stochastic neighbor embedding improve visualization and analysis of large datasets . Nat Commun . 2019 Nov 28; 10 ( 1 ): 5415 . OpenUrl CrossRef PubMed 70. ↵ Demšar J , Demsar J , Curk T , Curk T , Erjavec A , Erjave A. Orange: Data Mining Toolbox in Python . 71. ↵ Demšar J , Leban G , Zupan B . FreeViz—An intelligent multivariate visualization approach to explorative analysis of biomedical data . Journal of Biomedical Informatics . 2007 Dec ; 40 ( 6 ): 661 – 71 . OpenUrl CrossRef PubMed 72. ↵ Pedregosa F , Varoquaux G , Gramfort A , Michel V , Thirion B , Grisel O , et al. Scikit-learn: Machine Learning in Python [Internet] . arXiv; 2018 [cited 2025 Feb 15]. Available from: http://arxiv.org/abs/1201.0490 73. ↵ Lundberg S , Lee SI. A Unified Approach to Interpreting Model Predictions [Internet] . arXiv; 2017 [cited 2025 Feb 15]. Available from: https://arxiv.org/abs/1705.07874 74. ↵ Uttley J. Power Analysis, Sample Size , and Assessment of Statistical Assumptions—Improving the Evidential Value of Lighting Research . LEUKOS . 2019 Jul 3; 15 ( 2–3 ): 143 – 62 . OpenUrl CrossRef 75. ↵ Rey D , Neuhäuser M. Wilcoxon-Signed-Rank Test . In: Lovric M , editor. International Encyclopedia of Statistical Science [Internet] . Berlin, Heidelberg: Springer Berlin Heidelberg ; 2011 [cited 2025 Feb 15]. p. 1658–9. Available from: http://link.springer.com/10.1007/978-3-642-04898-2_616 76. ↵ Nanda A , Mohapatra DrBB , Mahapatra APK , Mahapatra APK , Mahapatra APK . Multiple comparison test by Tukey’s honestly significant difference (HSD): Do the confident level control type I error . Int J Stat Appl Math . 2021 Jan 1; 6 ( 1 ): 59 – 65 . OpenUrl CrossRef View the discussion thread. Back to top Previous Next Posted April 16, 2025. Download PDF Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following InteracTor: Feature Engineering and Explainable AI for Profiling Protein Structure-Interaction-Function Relationships Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share InteracTor: Feature Engineering and Explainable AI for Profiling Protein Structure-Interaction-Function Relationships Jose Cleydson F. Silva , Layla Schuster , Nick Sexson , Melissa Erdem , Ryan Hulk , Matias Kirst , Marcio F. R. Resende , Raquel Dias bioRxiv 2025.04.10.648139; doi: https://doi.org/10.1101/2025.04.10.648139 Share This Article: Copy Citation Tools InteracTor: Feature Engineering and Explainable AI for Profiling Protein Structure-Interaction-Function Relationships Jose Cleydson F. Silva , Layla Schuster , Nick Sexson , Melissa Erdem , Ryan Hulk , Matias Kirst , Marcio F. R. Resende , Raquel Dias bioRxiv 2025.04.10.648139; doi: https://doi.org/10.1101/2025.04.10.648139 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Bioinformatics Subject Areas All Articles Animal Behavior and Cognition (7635) Biochemistry (17697) Bioengineering (13895) Bioinformatics (41951) Biophysics (21456) Cancer Biology (18594) Cell Biology (25520) Clinical Trials (138) Developmental Biology (13381) Ecology (19903) Epidemiology (2067) Evolutionary Biology (24323) Genetics (15612) Genomics (22510) Immunology (17738) Microbiology (40401) Molecular Biology (17184) Neuroscience (88622) Paleontology (667) Pathology (2833) Pharmacology and Toxicology (4825) Physiology (7644) Plant Biology (15158) Scientific Communication and Education (2046) Synthetic Biology (4296) Systems Biology (9825) Zoology (2271)","source_license":"CC-BY-4.0","license_restricted":false}