Using computational approaches to enhance the interpretation of missense variants in the PAX6 gene

The PAX6 gene encodes a highly-conserved transcription factor involved in eye development. Heterozygous loss-of-function variants in PAX6 can cause a range of ophthalmic disorders including aniridia. A key molecular diagnostic challenge is that many PAX6 missense changes are presently classified as variants of uncertain significance. While computational tools can be used to assess the effect of genetic alterations, the accuracy of their predictions varies. Here, we evaluated and optimised the performance of computational prediction tools in relation to PAX6 missense variants. Through inspection of publicly available resources (including HGMD, ClinVar, LOVD and gnomAD), we identified 241 PAX6 missense variants that were used for model training and evaluation. The performance of ten commonly used computational tools was assessed and a threshold optimization approach was utilized to determine optimal cut-off values. Validation studies were subsequently undertaken using PAX6 variants from a local database. AlphaMissense, SIFT4G and REVEL emerged as the best-performing predictors; the optimized thresholds of these tools were 0.967, 0.025, and 0.772, respectively. Combining the prediction from these top-three tools resulted in lower performance compared to using AlphaMissense alone. Tailoring the use of computational tools by employing optimized thresholds specific to PAX6 can enhance algorithmic performance. Our findings have implications for PAX6 variant interpretation in clinical settings.

REEV: review, evaluate and explain variants

In the era of high throughput sequencing, special software is required for the clinical evaluation of genetic variants. We developed REEV (Review, Evaluate and Explain Variants), a user-friendly platform for clinicians and researchers in the field of rare disease genetics. Supporting data was aggregated from public data sources. We compared REEV with seven other tools for clinical variant evaluation. REEV (semi-)automatically fills individual ACMG criteria facilitating variant interpretation. REEV can store disease and phenotype data related to a case to use these for phenotype similarity measures. Users can create public permanent links for individual variants that can be saved as browser bookmarks and shared. REEV may help in the fast diagnostic assessment of genetic variants in a clinical as well as in a research context. REEV (https://reev.bihealth.org/) is free and open to all users and there is no login requirement.

Structure-based network analysis predicts pathogenic variants in human proteins associated with inherited retinal disease

Advances in gene sequencing technologies have accelerated the identification of genetic variants, but better tools are needed to understand which are causal of disease. This would be particularly useful in fields where gene therapy is a potential therapeutic modality for a disease-causing variant such as inherited retinal disease (IRD). Here, we apply structure-based network analysis (SBNA), which has been successfully utilized to identify variant-constrained amino acid residues in viral proteins, to identify residues that may cause IRD if subject to missense mutation. SBNA is based entirely on structural first principles and is not fit to specific outcome data, which makes it distinct from other contemporary missense prediction tools. In 4 well-studied human disease-associated proteins (BRCA1, HRAS, PTEN, and ERK2) with high-quality structural data, we find that SBNA scores correlate strongly with deep mutagenesis data. When applied to 47 IRD genes with available high-quality crystal structure data, SBNA scores reliably identified disease-causing variants according to phenotype definitions from the ClinVar database. Finally, we applied this approach to 63 patients at Massachusetts Eye and Ear (MEE) with IRD but for whom no genetic cause had been identified. Untrained models built using SBNA scores and BLOSUM62 scores for IRD-associated genes successfully predicted the pathogenicity of novel variants (AUC = 0.851), allowing us to identify likely causative disease variants in 40 IRD patients. Model performance was further augmented by incorporating orthogonal data from EVE scores (AUC = 0.927), which are based on evolutionary multiple sequence alignments. In conclusion, SBNA can used to successfully identify variants as causal of disease in human proteins and may help predict variants causative of IRD in an unbiased fashion.

CADD v1.7: using protein language models, regulatory CNNs and other nucleotide-level scores to improve genome-wide variant predictions

Machine Learning-based scoring and classification of genetic variants aids the assessment of clinical findings and is employed to prioritize variants in diverse genetic studies and analyses. Combined Annotation-Dependent Depletion (CADD) is one of the first methods for the genome-wide prioritization of variants across different molecular functions and has been continuously developed and improved since its original publication. Here, we present our most recent release, CADD v1.7. We explored and integrated new annotation features, among them state-of-the-art protein language model scores (Meta ESM-1v), regulatory variant effect predictions (from sequence-based convolutional neural networks) and sequence conservation scores (Zoonomia). We evaluated the new version on data sets derived from ClinVar, ExAC/gnomAD and 1000 Genomes variants. For coding effects, we tested CADD on 31 Deep Mutational Scanning (DMS) data sets from ProteinGym and, for regulatory effect prediction, we used saturation mutagenesis reporter assay data of promoter and enhancer sequences. The inclusion of new features further improved the overall performance of CADD. As with previous releases, all data sets, genome-wide CADD v1.7 scores, scripts for on-site scoring and an easy-to-use webserver are readily provided via https://cadd.bihealth.org/ or https://cadd.gs.washington.edu/ to the community.

Genome interpretation in a federated learning context allows the multi-center exome-based risk prediction of Crohn's disease patients

High-throughput sequencing allowed the discovery of many disease variants, but nowadays it is becoming clear that the abundance of genomics data mostly just moved the bottleneck in Genetics and Precision Medicine from a data availability issue to a data interpretation issue. To solve this empasse it would be beneficial to apply the latest Deep Learning (DL) methods to the Genome Interpretation (GI) problem, similarly to what AlphaFold did for Structural Biology. Unfortunately DL requires large datasets to be viable, and aggregating genomics datasets poses several legal, ethical and infrastructural complications. Federated Learning (FL) is a Machine Learning (ML) paradigm designed to tackle these issues. It allows ML methods to be collaboratively trained and tested on collections of physically separate datasets, without requiring the actual centralization of sensitive data. FL could thus be key to enable DL applications to GI on sufficiently large genomics data. We propose FedCrohn, a FL GI Neural Network model for the exome-based Crohn's Disease risk prediction, providing a proof-of-concept that FL is a viable paradigm to build novel ML GI approaches. We benchmark it in several realistic scenarios, showing that FL can indeed provide performances similar to conventional ML on centralized data, and that collaborating in FL initiatives is likely beneficial for most of the medical centers participating in them.

Structure-based network analysis predicts mutations associated with inherited retinal disease

With continued advances in gene sequencing technologies comes the need to develop better tools to understand which mutations cause disease. Here we validate structure-based network analysis (SBNA)1,2 in well-studied human proteins and report results of using SBNA to identify critical amino acids that may cause retinal disease if subject to missense mutation. We computed SBNA scores for genes with high-quality structural data, starting with validating the method using 4 well-studied human disease-associated proteins. We then analyzed 47 inherited retinal disease (IRD) genes. We compared SBNA scores to phenotype data from the ClinVar database and found a significant difference between benign and pathogenic mutations with respect to network score. Finally, we applied this approach to 65 patients at Massachusetts Eye and Ear (MEE) who were diagnosed with IRD but for whom no genetic cause was found. Multivariable logistic regression models built using SBNA scores for IRD-associated genes successfully predicted pathogenicity of novel mutations, allowing us to identify likely causative disease variants in 37 patients with IRD from our clinic. In conclusion, SBNA can be meaningfully applied to human proteins and may help predict mutations causative of IRD.

Analyses of Mutation Displacements from Homology Models

Evaluation of the structural perturbations introduced by a single amino acid mutation is the main issue for protein structural biology. We propose here to present some recent advances in methods, allowing the splitting of distortion between the actual substitution effect and the contribution of the local flexibility of the position where the mutation occurs. Its main drawback is the need of many structures with a single mutation in each of them. To bypass this difficulty, we propose to use molecular modeling tools, with several software enabling us to build a model from a template, given the sequence. As a proof of concept, we rely on a gold standard, the human lysozyme. Both wild-type and three mutant structures are available in the PDB. Two of these mutations result in amyloid fibril formation, and the last one is neutral. As a conclusion, irrespective of the algorithm used for modeling, side chain conformations at the site of mutation are reliable, although long-range effects are out of reach of these tools.

ProTstab2 for Prediction of Protein Thermal Stabilities

The stability of proteins is an essential property that has several biological implications. Knowledge about protein stability is important in many ways, ranging from protein purification and structure determination to stability in cells and biotechnological applications. Experimental determination of thermal stabilities has been tedious and available data have been limited. The introduction of limited proteolysis and mass spectrometry approaches has facilitated more extensive cellular protein stability data production. We collected melting temperature information for 34,913 proteins and developed a machine learning predictor, ProTstab2, by utilizing a gradient boosting algorithm after testing seven algorithms. The method performance was assessed on a blind test data set and showed a Pearson correlation coefficient of 0.753 and root mean square error of 7.005. Comparison to previous methods indicated that ProTstab2 had superior performance. The method is fast, so it was applied to predict and compare the stabilities of all proteins in human, mouse, and zebrafish proteomes for which experimental data were not determined. The tool is freely available.

Vasor: Accurate prediction of variant effects for amino acid substitutions in multidrug resistance protein 3

The phosphatidylcholine floppase multidrug resistance protein 3 (MDR3) is an essential hepatobiliary transport protein. MDR3 dysfunction is associated with various liver diseases, ranging from severe progressive familial intrahepatic cholestasis to transient forms of intrahepatic cholestasis of pregnancy and familial gallstone disease. Single amino acid substitutions are often found as causative of dysfunction, but identifying the substitution effect in in vitro studies is time and cost intensive. We developed variant assessor of MDR3 (Vasor), a machine learning-based model to classify novel MDR3 missense variants into the categories benign or pathogenic. Vasor was trained on the largest data set to date that is specific for benign and pathogenic variants of MDR3 and uses general predictors, namely Evolutionary Models of Variant Effects (EVE), EVmutation, PolyPhen-2, I-Mutant2.0, MUpro, MAESTRO, and PON-P2 along with other variant properties, such as half-sphere exposure and posttranslational modification site, as input. Vasor consistently outperformed the integrated general predictors and the external prediction tool MutPred2, leading to the current best prediction performance for MDR3 single-site missense variants (on an external test set: F1-score, 0.90; Matthew's correlation coefficient, 0.80). Furthermore, Vasor predictions cover the entire sequence space of MDR3. Vasor is accessible as a webserver at https://cpclab.uni-duesseldorf.de/mdr3_predictor/ for users to rapidly obtain prediction results and a visualization of the substitution site within the MDR3 structure. The MDR3-specific prediction tool Vasor can provide reliable predictions of single-site amino acid substitutions, giving users a fast way to initially assess whether a variant is benign or pathogenic.

Individual Genetic Heterogeneity

Genetic variation has been widely covered in literature, however, not from the perspective of an individual in any species. Here, a synthesis of genetic concepts and variations relevant for individual genetic constitution is provided. All the different levels of genetic information and variation are covered, ranging from whether an organism is unmixed or hybrid, has variations in genome, chromosomes, and more locally in DNA regions, to epigenetic variants or alterations in selfish genetic elements. Genetic constitution and heterogeneity of microbiota are highly relevant for health and wellbeing of an individual. Mutation rates vary widely for variation types, e.g., due to the sequence context. Genetic information guides numerous aspects in organisms. Types of inheritance, whether Mendelian or non-Mendelian, zygosity, sexual reproduction, and sex determination are covered. Functions of DNA and functional effects of variations are introduced, along with mechanism that reduce and modulate functional effects, including TARAR countermeasures and intraindividual genetic conflict. TARAR countermeasures for tolerance, avoidance, repair, attenuation, and resistance are essential for life, integrity of genetic information, and gene expression. The genetic composition, effects of variations, and their expression are considered also in diseases and personalized medicine. The text synthesizes knowledge and insight on individual genetic heterogeneity and organizes and systematizes the central concepts.

A Comprehensive Evaluation of the Performance of Prediction Algorithms on Clinically Relevant Missense Variants

The rapid integration of genomic technologies in clinical diagnostics has resulted in the detection of a multitude of missense variants whose clinical significance is often unknown. As a result, a plethora of computational tools have been developed to facilitate variant interpretation. However, choosing an appropriate software from such a broad range of tools can be challenging; therefore, systematic benchmarking with high-quality, independent datasets is critical. Using three independent benchmarking datasets compiled from the ClinVar database, we evaluated the performance of ten widely used prediction algorithms with missense variants from 21 clinically relevant genes, including BRCA1 and BRCA2. A fourth dataset consisting of 1053 missense variants was also used to investigate the impact of type 1 circularity on their performance. The performance of the prediction algorithms varied widely across datasets. Based on Matthews Correlation Coefficient and Area Under the Curve, SNPs&GO and PMut consistently displayed an overall above-average performance across the datasets. Most of the tools demonstrated greater sensitivity and negative predictive values at the expense of lower specificity and positive predictive values. We also demonstrated that type 1 circularity significantly impacts the performance of these tools and, if not accounted for, may confound the selection of the best performing algorithms.

How Functional Genomics Can Keep Pace With VUS Identification

Over the last two decades, an exponentially expanding number of genetic variants have been identified associated with inherited cardiac conditions. These tremendous gains also present challenges in deciphering the clinical relevance of unclassified variants or variants of uncertain significance (VUS). This review provides an overview of the advancements (and challenges) in functional and computational approaches to characterize variants and help keep pace with VUS identification related to inherited heart diseases.

PON-All: Amino Acid Substitution Tolerance Predictor for All Organisms

Genetic variations are investigated in human and many other organisms for many purposes (e.g., to aid in clinical diagnosis). Interpretation of the identified variations can be challenging. Although some dedicated prediction methods have been developed and some tools for human variants can also be used for other organisms, the performance and species range have been limited. We developed a novel variant pathogenicity/tolerance predictor for amino acid substitutions in any organism. The method, PON-All, is a machine learning tool trained on human, animal, and plant variants. Two versions are provided, one with Gene Ontology (GO) annotations and another without these details. GO annotations are not available or are partial for many organisms of interest. The methods provide predictions for three classes: pathogenic, benign, and variants of unknown significance. On the blind test, when using GO annotations, accuracy was 0.913 and MCC 0.827. When GO features were not used, accuracy was 0.856 and MCC 0.712. The performance is the best for human and plant variants and somewhat lower for animal variants because the number of known disease-causing variants in animals is rather small. The method was compared to several other tools and was found to have superior performance. PON-All is freely available at http://structure.bmc.lu.se/PON-All and http://8.133.174.28:8999/.

Integration of Protein Structure and Population-Scale DNA Sequence Data for Disease Gene Discovery and Variant Interpretation

The experimental and computational techniques for capturing information about protein structures and genetic variation within the human genome have advanced dramatically in the past 20 years, generating extensive new data resources. In this review, we discuss these advances, along with new approaches for determining the impact a genetic variant has on protein function. We focus on the potential of new methods that integrate human genetic variation into protein structures to discover relationships to disease, including the discovery of mutational hotspots in cancer-related proteins, the localization of protein-altering variants within protein regions for common complex diseases, and the assessment of variants of unknown significance for Mendelian traits. We expect that approaches that integrate these data sources will play increasingly important roles in disease gene discovery and variant interpretation. Expected final online publication date for the Annual Review of Biomedical Data Science, Volume 5 is August 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Data set and fitting dependencies when estimating protein mutant stability: Toward simple, balanced, and interpretable models

Accurate prediction of protein stability changes upon mutation (ΔΔG) is increasingly important to evolution studies, protein engineering, and screening of disease-causing gene variants but is challenged by biases in training data. We investigated 45 linear regression models trained on data sets that account systematically for destabilization bias and mutation-type bias B_M . The models were externally validated on three test data sets probing different pathologies and for internal consistency (symmetry and neutrality). Model structure and performance substantially depended on training data and even fitting method. We developed two final models: SimBa-IB for typical natural mutations and SimBa-SYM for situations where stabilizing and destabilizing mutations occur to a similar extent. SimBa-SYM, despite is simplicity, is essentially non-biased (vs. the S^sym data set) while still performing well for all data sets (R ~ 0.46-0.54, MAE = 1.16-1.24 kcal/mol). The simple models provide advantage in terms of interpretability, use and future improvement, and are freely available on GitHub.

Identification and in silico Characterization of Deleterious Single Nucleotide Variations in Human ZP2 Gene

ZP2, an important component of the zona matrix, surrounds mammalian oocytes and facilitates fertilization. Recently, some studies have documented the association of mutations in genes encoding the zona matrix with the infertile status of human females. Single nucleotide polymorphisms are the most common type of genetic variations observed in a population and as per the dbSNP database, around 5,152 SNPs are reported to exist in the human ZP2 (hZP2) gene. Although a wide range of computational tools are publicly available, yet no computational studies have been done to date to identify and analyze structural and functional effects of deleterious SNPs on hZP2. In this study, we conducted a comprehensive in silico analysis of all the SNPs found in hZP2. Six different computational tools including SIFT and PolyPhen-2 predicted 18 common nsSNPs as deleterious of which 12 were predicted to most likely affect the structure/functional properties. These were either present in the N-term region crucial for sperm-zona interaction or in the zona domain. 31 additional SNPs in both coding and non-coding regions were also identified. Interestingly, some of these SNPs have been found to be present in infertile females in some recent studies.

From genotype to phenotype in Arabidopsis thaliana: in-silico genome interpretation predicts 288 phenotypes from sequencing data

In many cases, the unprecedented availability of data provided by high-throughput sequencing has shifted the bottleneck from a data availability issue to a data interpretation issue, thus delaying the promised breakthroughs in genetics and precision medicine, for what concerns Human genetics, and phenotype prediction to improve plant adaptation to climate change and resistance to bioagressors, for what concerns plant sciences. In this paper, we propose a novel Genome Interpretation paradigm, which aims at directly modeling the genotype-to-phenotype relationship, and we focus on A. thaliana since it is the best studied model organism in plant genetics. Our model, called Galiana, is the first end-to-end Neural Network (NN) approach following the genomes in/phenotypes out paradigm and it is trained to predict 288 real-valued Arabidopsis thaliana phenotypes from Whole Genome sequencing data. We show that 75 of these phenotypes are predicted with a Pearson correlation ≥0.4, and are mostly related to flowering traits. We show that our end-to-end NN approach achieves better performances and larger phenotype coverage than models predicting single phenotypes from the GWAS-derived known associated genes. Galiana is also fully interpretable, thanks to the Saliency Maps gradient-based approaches. We followed this interpretation approach to identify 36 novel genes that are likely to be associated with flowering traits, finding evidence for 6 of them in the existing literature.

Accurate assignment of disease liability to genetic variants using only population data

PURPOSE

The growing size of public variant repositories prompted us to test the accuracy of pathogenicity prediction of DNA variants using population data alone.

METHODS

Under the a priori assumption that the ratio of the prevalence of variants in healthy population vs that in affected populations form 2 distinct distributions (pathogenic and benign), we used a Bayesian method to assign probability to a variant belonging to either distribution.

RESULTS

The approach, termed Bayesian prevalence ratio (BayPR), accurately parsed 300 of 313 expertly curated CFTR variants: 284 of 296 pathogenic/likely pathogenic variants in 1 distribution and 16 of 17 benign/likely benign variants in another. BayPR produced an area under the receiver operating characteristic curve of 0.99 for 103 functionally confirmed missense CFTR variants, which is equal to or exceeds 10 commonly used algorithms (area under the receiver operating characteristic curve range = 0.54-0.99). Application of BayPR to expertly curated variants in 8 genes associated with 7 Mendelian conditions led to the assignment of a disease-causing probability of ≥80% to 1350 of 1374 (98.3%) pathogenic/likely pathogenic variants and of ≤20% to 22 of 23 (95.7%) benign/likely benign variants.

CONCLUSION

Irrespective of the variant type or functional effect, the BayPR approach provides probabilities of pathogenicity for DNA variants responsible for Mendelian disorders using only the variant counts in affected and unaffected population samples.

L-Type Calcium Channel: Predicting Pathogenic/Likely Pathogenic Status for Variants of Uncertain Clinical Significance

(1) Background: Defects in gene CACNA1C, which encodes the pore-forming subunit of the human Cav1.2 channel (hCav1.2), are associated with cardiac disorders such as atrial fibrillation, long QT syndrome, conduction disorders, cardiomyopathies, and congenital heart defects. Clinical manifestations are known only for 12% of CACNA1C missense variants, which are listed in public databases. Bioinformatics approaches can be used to predict the pathogenic/likely pathogenic status for variants of uncertain clinical significance. Choosing a bioinformatics tool and pathogenicity threshold that are optimal for specific protein families increases the reliability of such predictions. (2) Methods and Results: We used databases ClinVar, Humsavar, gnomAD, and Ensembl to compose a dataset of pathogenic/likely pathogenic and benign variants of hCav1.2 and its 20 paralogues: voltage-gated sodium and calcium channels. We further tested the performance of sixteen in silico tools in predicting pathogenic variants. ClinPred demonstrated the best performance, followed by REVEL and MCap. In the subset of 309 uncharacterized variants of hCav1.2, ClinPred predicted the pathogenicity for 188 variants. Among these, 36 variants were also categorized as pathogenic/likely pathogenic in at least one paralogue of hCav1.2. (3) Conclusions: The bioinformatics tool ClinPred and the paralogue annotation method consensually predicted the pathogenic/likely pathogenic status for 36 uncharacterized variants of hCav1.2. An analogous approach can be used to classify missense variants of other calcium channels and novel variants of hCav1.2.

Analysis and Interpretation of the Impact of Missense Variants in Cancer

Large scale genome sequencing allowed the identification of a massive number of genetic variations, whose impact on human health is still unknown. In this review we analyze, by an in silico-based strategy, the impact of missense variants on cancer-related genes, whose effect on protein stability and function was experimentally determined. We collected a set of 164 variants from 11 proteins to analyze the impact of missense mutations at structural and functional levels, and to assess the performance of state-of-the-art methods (FoldX and Meta-SNP) for predicting protein stability change and pathogenicity. The result of our analysis shows that a combination of experimental data on protein stability and in silico pathogenicity predictions allowed the identification of a subset of variants with a high probability of having a deleterious phenotypic effect, as confirmed by the significant enrichment of the subset in variants annotated in the COSMIC database as putative cancer-driving variants. Our analysis suggests that the integration of experimental and computational approaches may contribute to evaluate the risk for complex disorders and develop more effective treatment strategies.

KEAP1 Cancer Mutants: A Large-Scale Molecular Dynamics Study of Protein Stability

We have performed 280 μs of unbiased molecular dynamics (MD) simulations to investigate the effects of 12 different cancer mutations on Kelch-like ECH-associated protein 1 (KEAP1) (G333C, G350S, G364C, G379D, R413L, R415G, A427V, G430C, R470C, R470H, R470S and G476R), one of the frequently mutated proteins in lung cancer. The aim was to provide structural insight into the effects of these mutants, including a new class of ANCHOR (additionally NRF2-complexed hypomorph) mutant variants. Our work provides additional insight into the structural dynamics of mutants that could not be analyzed experimentally, painting a more complete picture of their mutagenic effects. Notably, blade-wise analysis of the Kelch domain points to stability as a possible target of cancer in KEAP1. Interestingly, structural analysis of the R470C ANCHOR mutant, the most prevalent missense mutation in KEAP1, revealed no significant change in structural stability or NRF2 binding site dynamics, possibly indicating an covalent modification as this mutant's mode of action.

Mapping OMIM Disease-Related Variations on Protein Domains Reveals an Association Among Variation Type, Pfam Models, and Disease Classes

Human genome resequencing projects provide an unprecedented amount of data about single-nucleotide variations occurring in protein-coding regions and often leading to observable changes in the covalent structure of gene products. For many of these variations, links to Online Mendelian Inheritance in Man (OMIM) genetic diseases are available and are reported in many databases that are collecting human variation data such as Humsavar. However, the current knowledge on the molecular mechanisms that are leading to diseases is, in many cases, still limited. For understanding the complex mechanisms behind disease insurgence, the identification of putative models, when considering the protein structure and chemico-physical features of the variations, can be useful in many contexts, including early diagnosis and prognosis. In this study, we investigate the occurrence and distribution of human disease–related variations in the context of Pfam domains. The aim of this study is the identification and characterization of Pfam domains that are statistically more likely to be associated with disease-related variations. The study takes into consideration 2,513 human protein sequences with 22,763 disease-related variations. We describe patterns of disease-related variation types in biunivocal relation with Pfam domains, which are likely to be possible markers for linking Pfam domains to OMIM diseases. Furthermore, we take advantage of the specific association between disease-related variation types and Pfam domains for clustering diseases according to the Human Disease Ontology, and we establish a relation among variation types, Pfam domains, and disease classes. We find that Pfam models are specific markers of patterns of variation types and that they can serve to bridge genes, diseases, and disease classes. Data are available as Supplementary Material for 1,670 Pfam models, including 22,763 disease-related variations associated to 3,257 OMIM diseases.

IDRMutPred: predicting disease-associated germline nonsynonymous single nucleotide variants (nsSNVs) in intrinsically disordered regions

Motivation

Despite of the lack of folded structure, intrinsically disordered regions (IDRs) of proteins play versatile roles in various biological processes, and many nonsynonymous single nucleotide variants (nsSNVs) in IDRs are associated with human diseases. The continuous accumulation of nsSNVs resulted from the wide application of NGS has driven the development of disease-association prediction methods for decades. However, their performance on nsSNVs in IDRs remains inferior, possibly due to the domination of nsSNVs from structured regions in training data. Therefore, it is highly demanding to build a disease-association predictor specifically for nsSNVs in IDRs with better performance.

Results

We present IDRMutPred, a machine learning-based tool specifically for predicting disease-associated germline nsSNVs in IDRs. Based on 17 selected optimal features that are extracted from sequence alignments, protein annotations, hydrophobicity indices and disorder scores, IDRMutPred was trained using three ensemble learning algorithms on the training dataset containing only IDR nsSNVs. The evaluation on the two testing datasets shows that all the three prediction models outperform 17 other popular general predictors significantly, achieving the ACC between 0.856 and 0.868 and MCC between 0.713 and 0.737. IDRMutPred will prioritize disease-associated IDR germline nsSNVs more reliably than general predictors.

Availability and implementation

The software is freely available at http://www.wdspdb.com/IDRMutPred.

Supplementary information

Supplementary data are available at Bioinformatics online.

The DBSAV Database: Predicting Deleteriousness of Single Amino Acid Variations in the Human Proteome

Deleterious single amino acid variation (SAV) is one of the leading causes of human diseases. Evaluating the functional impact of SAVs is crucial for diagnosis of genetic disorders. We previously developed a deep convolutional neural network predictor, DeepSAV, to evaluate the deleterious effects of SAVs on protein function based on various sequence, structural, and functional properties. DeepSAV scores of rare SAVs observed in the human population are aggregated into a gene-level score called GTS (Gene Tolerance of rare SAVs) that reflects a gene's tolerance to deleterious missense mutations and serves as a useful tool to study gene-disease associations. In this study, we aim to enhance the performance of DeepSAV by using expanded datasets of pathogenic and benign variants, more features, and neural network optimization. We found that multiple sequence alignments built from vertebrate-level orthologs yield better prediction results compared to those built from mammalian-level orthologs. For multiple sequence alignments built from BLAST searches, optimal performance was achieved with a sequence identify cutoff of 50% to remove distant homologs. The new version of DeepSAV exhibits the best performance among standalone predictors of deleterious effects of SAVs. We developed the DBSAV database (http://prodata.swmed.edu/DBSAV) that reports GTS scores of human genes and DeepSAV scores of SAVs in the human proteome, including pathogenic and benign SAVs, population-level SAVs, and all possible SAVs by single nucleotide variations. This database serves as a useful resource for research of human SAVs and their relationships with protein functions and human diseases.

TMSNP: a web server to predict pathogenesis of missense mutations in the transmembrane region of membrane proteins

The massive amount of data generated from genome sequencing brings tons of newly identified mutations, whose pathogenic/non-pathogenic effects need to be evaluated. This has given rise to several mutation predictor tools that, in general, do not consider the specificities of the various protein groups. We aimed to develop a predictor tool dedicated to membrane proteins, under the premise that their specific structural features and environment would give different responses to mutations compared to globular proteins. For this purpose, we created TMSNP, a database that currently contains information from 2624 pathogenic and 196 705 non-pathogenic reported mutations located in the transmembrane region of membrane proteins. By computing various conservation parameters on these mutations in combination with annotations, we trained a machine-learning model able to classify mutations as pathogenic or not. TMSNP (freely available at http://lmc.uab.es/tmsnp/) improves considerably the prediction power of commonly used mutation predictors trained with globular proteins.

A base measure of precision for protein stability predictors: structural sensitivity

BACKGROUND

Prediction of the change in fold stability (ΔΔG) of a protein upon mutation is of major importance to protein engineering and screening of disease-causing variants. Many prediction methods can use 3D structural information to predict ΔΔG. While the performance of these methods has been extensively studied, a new problem has arisen due to the abundance of crystal structures: How precise are these methods in terms of structure input used, which structure should be used, and how much does it matter? Thus, there is a need to quantify the structural sensitivity of protein stability prediction methods.

RESULTS

We computed the structural sensitivity of six widely-used prediction methods by use of saturated computational mutagenesis on a diverse set of 87 structures of 25 proteins. Our results show that structural sensitivity varies massively and surprisingly falls into two very distinct groups, with methods that take detailed account of the local environment showing a sensitivity of ~ 0.6 to 0.8 kcal/mol, whereas machine-learning methods display much lower sensitivity (~ 0.1 kcal/mol). We also observe that the precision correlates with the accuracy for mutation-type-balanced data sets but not generally reported accuracy of the methods, indicating the importance of mutation-type balance in both contexts.

CONCLUSIONS

The structural sensitivity of stability prediction methods varies greatly and is caused mainly by the models and less by the actual protein structural differences. As a new recommended standard, we therefore suggest that ΔΔG values are evaluated on three protein structures when available and the associated standard deviation reported, to emphasize not just the accuracy but also the precision of the method in a specific study. Our observation that machine-learning methods deemphasize structure may indicate that folded wild-type structures alone, without the folded mutant and unfolded structures, only add modest value for assessing protein stability effects, and that side-chain-sensitive methods overstate the significance of the folded wild-type structure.

Alignment-free method for functional annotation of amino acid substitutions: Application on epigenetic factors involved in hematologic malignancies

For the last couple of decades, there has been a significant growth in sequencing data, leading to an extraordinary increase in the number of gene variants. This places a challenge on the bioinformatics research community to develop and improve computational tools for functional annotation of new variants. Genes coding for epigenetic regulators have important roles in cancer pathogenesis and mutations in these genes show great potential as clinical biomarkers, especially in hematologic malignancies. Therefore, we developed a model that specifically focuses on these genes, with an assumption that it would outperform general models in predicting the functional effects of amino acid substitutions. EpiMut is a standalone software that implements a sequence based alignment-free method. We applied a two-step approach for generating sequence based features, relying on the biophysical and biochemical indices of amino acids and the Fourier Transform as a sequence transformation method. For each gene in the dataset, the machine learning algorithm-Naïve Bayes was used for building a model for prediction of the neutral or disease-related status of variants. EpiMut outperformed state-of-the-art tools used for comparison, PolyPhen-2, SIFT and SNAP2. Additionally, EpiMut showed the highest performance on the subset of variants positioned outside conserved functional domains of analysed proteins, which represents an important group of cancer-related variants. These results imply that EpiMut can be applied as a first choice tool in research of the impact of gene variants in epigenetic regulators, especially in the light of the biomarker role in hematologic malignancies. EpiMut is freely available at https://www.vin.bg.ac.rs/180/tools/epimut.php.

Clinical profiles and diagnostic challenges in 1158 children with rare hepatobiliary disorders

BACKGROUND

Diagnosis of rare diseases possesses a great challenge in pediatric hepatology because expert knowledge in the field is extremely insufficient. The study aims to explore new findings and collect diagnostic experience from pediatric rare liver diseases.

METHODS

The large-sample case analysis study included pediatric patients who had liver-involved rare diseases. All cases underwent liver biopsy and/or gene sequencing.

RESULTS

A total of 1158 pediatric patients were identified. Liver-based genetic diseases were most frequent (737 cases), followed by liver damages involved in extrahepatic or systemic disorders (151 cases) and cryptogenic hepatobilliary abnormalities (123 cases). Of note, diagnoses of 16 patients were re-evaluated according to genetic results combined with clinical pointers. In addition, 101 patients who underwent gene sequencing remained undiagnosed. Of them, 55 had negative genetic findings, 30 harbored mutations that failed to meet their typically pathogenic condition, and 16 had detected variants that were inconsistent with clinical pointers.

CONCLUSIONS

As a study involving known largest number of children with rare hepatobiliary disorders, it allows us to accumulate information (especially new findings) on the etiology and diagnosis of these disorders. The results can help to improve the diagnostic quality in the population.

IMPACT

Liver-based genetic diseases were most frequent in clinical profiles of pediatric rare liver diseases. Some novel variants in cases with genetic diseases (for example, two variants of c.3638G>T and c.1435G>C in a patient with progressive familial intrahepatic cholestasis type 2) were identified. As a study involving known largest number of pediatric cases with rare hepatobiliary disorders, it allows us to accumulate information on the etiology and diagnosis of these disorders. The study can help to optimize the diagnostic process and significantly improve the diagnostic quality in the field of pediatric hepatology. Given that clinical variability often exists within rare genetic disease entities and not all rare disorders are genetic, clinicians should not over-depend on the genetic results in the diagnosis.

Inferring the molecular and phenotypic impact of amino acid variants with MutPred2

Identifying pathogenic variants and underlying functional alterations is challenging. To this end, we introduce MutPred2, a tool that improves the prioritization of pathogenic amino acid substitutions over existing methods, generates molecular mechanisms potentially causative of disease, and returns interpretable pathogenicity score distributions on individual genomes. Whilst its prioritization performance is state-of-the-art, a distinguishing feature of MutPred2 is the probabilistic modeling of variant impact on specific aspects of protein structure and function that can serve to guide experimental studies of phenotype-altering variants. We demonstrate the utility of MutPred2 in the identification of the structural and functional mutational signatures relevant to Mendelian disorders and the prioritization of de novo mutations associated with complex neurodevelopmental disorders. We then experimentally validate the functional impact of several variants identified in patients with such disorders. We argue that mechanism-driven studies of human inherited disease have the potential to significantly accelerate the discovery of clinically actionable variants.

Variation benchmark datasets: update, criteria, quality and applications

Development of new computational methods and testing their performance has to be carried out using experimental data. Only in comparison to existing knowledge can method performance be assessed. For that purpose, benchmark datasets with known and verified outcome are needed. High-quality benchmark datasets are valuable and may be difficult, laborious and time consuming to generate. VariBench and VariSNP are the two existing databases for sharing variation benchmark datasets used mainly for variation interpretation. They have been used for training and benchmarking predictors for various types of variations and their effects. VariBench was updated with 419 new datasets from 109 papers containing altogether 329 014 152 variants; however, there is plenty of redundancy between the datasets. VariBench is freely available at http://structure.bmc.lu.se/VariBench/. The contents of the datasets vary depending on information in the original source. The available datasets have been categorized into 20 groups and subgroups. There are datasets for insertions and deletions, substitutions in coding and non-coding region, structure mapped, synonymous and benign variants. Effect-specific datasets include DNA regulatory elements, RNA splicing, and protein property for aggregation, binding free energy, disorder and stability. Then there are several datasets for molecule-specific and disease-specific applications, as well as one dataset for variation phenotype effects. Variants are often described at three molecular levels (DNA, RNA and protein) and sometimes also at the protein structural level including relevant cross references and variant descriptions. The updated VariBench facilitates development and testing of new methods and comparison of obtained performances to previously published methods. We compared the performance of the pathogenicity/tolerance predictor PON-P2 to several benchmark studies, and show that such comparisons are feasible and useful, however, there may be limitations due to lack of provided details and shared data.

Database URL: http://structure.bmc.lu.se/VariBench

Functional effects of protein variants

Genetic and other variations frequently affect protein functions. Scientific articles can contain confusing descriptions about which function or property is affected, and in many cases the statements are pure speculation without any experimental evidence. To clarify functional effects of protein variations of genetic or non-genetic origin, a systematic conceptualisation and framework are introduced. This framework describes protein functional effects on abundance, activity, specificity and affinity, along with countermeasures, which allow cells, tissues and organisms to tolerate, avoid, repair, attenuate or resist (TARAR) the effects. Effects on abundance discussed include gene dosage, restricted expression, mis-localisation and degradation. Enzymopathies, effects on kinetics, allostery and regulation of protein activity are subtopics for the effects of variants on activity. Variation outcomes on specificity and affinity comprise promiscuity, specificity, affinity and moonlighting. TARAR mechanisms redress variations with active and passive processes including chaperones, redundancy, robustness, canalisation and metabolic and signalling rewiring. A framework for pragmatic protein function analysis and presentation is introduced. All of the mechanisms and effects are described along with representative examples, most often in relation to diseases. In addition, protein function is discussed from evolutionary point of view. Application of the presented framework facilitates unambiguous, detailed and specific description of functional effects and their systematic study.

Identification of pathogenic missense mutations using protein stability predictors

Attempts at using protein structures to identify disease-causing mutations have been dominated by the idea that most pathogenic mutations are disruptive at a structural level. Therefore, computational stability predictors, which assess whether a mutation is likely to be stabilising or destabilising to protein structure, have been commonly used when evaluating new candidate disease variants, despite not having been developed specifically for this purpose. We therefore tested 13 different stability predictors for their ability to discriminate between pathogenic and putatively benign missense variants. We find that one method, FoldX, significantly outperforms all other predictors in the identification of disease variants. Moreover, we demonstrate that employing predicted absolute energy change scores improves performance of nearly all predictors in distinguishing pathogenic from benign variants. Importantly, however, we observe that the utility of computational stability predictors is highly heterogeneous across different proteins, and that they are all inferior to the best performing variant effect predictors for identifying pathogenic mutations. We suggest that this is largely due to alternate molecular mechanisms other than protein destabilisation underlying many pathogenic mutations. Thus, better ways of incorporating protein structural information and molecular mechanisms into computational variant effect predictors will be required for improved disease variant prioritisation.

Novel F8 and F9 gene variants from the PedNet hemophilia registry classified according to ACMG/AMP guidelines

In hemophilia A and B, analysis of the F8 and F9 gene variants enables carrier and prenatal diagnosis and prediction of risk for the development of inhibitors. The PedNet Registry collects clinical, genetic, and phenotypic data prospectively on more than 2000 children with hemophilia. The genetic reports of F8/F9 gene variants were classified uniformly to Human Genome Variation Society nomenclature and reevaluated using international population‐ and disease‐specific databases, literature survey and, where applicable, computational predictive programs. We report 88 novel variants in the F8 and F9 genes, 80 fulfilling criteria for Class 5 (pathogenic), six for Class 4 (likely pathogenic) and two fulfilling criteria for Class 3 (variant of unknown significance) of the American College of Medical Genetics and Genomics/Association for Molecular Pathologyguidelines together with information on the respective phenotype and inhibitor formation. The study highlights the need to reevaluate and update earlier genetic reports in hemophilia both locally but also in variant databases in light of changed nomenclature and new guidelines.

Mutation severity spectrum of rare alleles in the human genome is predictive of disease type

The human genome harbors a variety of genetic variations. Single-nucleotide changes that alter amino acids in protein-coding regions are one of the major causes of human phenotypic variation and diseases. These single-amino acid variations (SAVs) are routinely found in whole genome and exome sequencing. Evaluating the functional impact of such genomic alterations is crucial for diagnosis of genetic disorders. We developed DeepSAV, a deep-learning convolutional neural network to differentiate disease-causing and benign SAVs based on a variety of protein sequence, structural and functional properties. Our method outperforms most stand-alone programs, and the version incorporating population and gene-level information (DeepSAV+PG) has similar predictive power as some of the best available. We transformed DeepSAV scores of rare SAVs in the human population into a quantity termed "mutation severity measure" for each human protein-coding gene. It reflects a gene's tolerance to deleterious missense mutations and serves as a useful tool to study gene-disease associations. Genes implicated in cancer, autism, and viral interaction are found by this measure as intolerant to mutations, while genes associated with a number of other diseases are scored as tolerant. Among known disease-associated genes, those that are mutation-intolerant are likely to function in development and signal transduction pathways, while those that are mutation-tolerant tend to encode metabolic and mitochondrial proteins.

Fido-SNP: the first webserver for scoring the impact of single nucleotide variants in the dog genome

As the amount of genomic variation data increases, tools that are able to score the functional impact of single nucleotide variants become more and more necessary. While there are several prediction servers available for interpreting the effects of variants in the human genome, only few have been developed for other species, and none were specifically designed for species of veterinary interest such as the dog. Here, we present Fido-SNP the first predictor able to discriminate between Pathogenic and Benign single-nucleotide variants in the dog genome. Fido-SNP is a binary classifier based on the Gradient Boosting algorithm. It is able to classify and score the impact of variants in both coding and non-coding regions based on sequence features within seconds. When validated on a previously unseen set of annotated variants from the OMIA database, Fido-SNP reaches 88% overall accuracy, 0.77 Matthews correlation coefficient and 0.91 Area Under the ROC Curve.

SAAMBE-3D: Predicting Effect of Mutations on Protein-Protein Interactions

Maintaining wild type protein-protein interactions is essential for the normal function of cell and any mutation that alter their characteristics can cause disease. Therefore, the ability to correctly and quickly predict the effect of amino acid mutations is crucial for understanding disease effects and to be able to carry out genome-wide studies. Here, we report a new development of the SAAMBE method, SAAMBE-3D, which is a machine learning-based approach, resulting in accurate predictions and is extremely fast. It achieves the Pearson correlation coefficient ranging from 0.78 to 0.82 depending on the training protocol in benchmarking five-fold validation test against the SKEMPI v2.0 database and outperforms currently existing algorithms on various blind-tests. Furthermore, optimized and tested via five-fold cross-validation on the Cornell University dataset, the SAAMBE-3D achieves AUC of 1.0 and 0.96 on a homo and hereto-dimer test datasets. Another important feature of SAAMBE-3D is that it is very fast, it takes less than a fraction of a second to complete a prediction. SAAMBE-3D is available as a web server and as well as a stand-alone code, the last one being another important feature allowing other researchers to directly download the code and run it on their local computer. Combined all together, SAAMBE-3D is an accurate and fast software applicable for genome-wide studies to assess the effect of amino acid mutations on protein-protein interactions. The webserver and the stand-alone codes (SAAMBE-3D for predicting the change of binding free energy and SAAMBE-3D-DN for predicting if the mutation is disruptive or non-disruptive) are available.

Problems in variation interpretation guidelines and in their implementation in computational tools

Background

ACMG/AMP and AMP/ASCO/CAP have released guidelines for variation interpretation, and ESHG for diagnostic sequencing. These guidelines contain recommendations including the use of computational prediction methods. The guidelines per se and the way they are implemented cause some problems.

Methods

Logical reasoning based on domain knowledge.

Results

According to the guidelines, several methods have to be used and they have to agree. This means that the methods with the poorest performance overrule the better ones. The choice of the prediction method(s) should be made by experts  based on systematic benchmarking studies reporting all the relevant performance measures. Currently variation interpretation methods have been applied mainly to amino acid substitutions and splice site variants; however, predictors for some other types of variations are available and there will be tools for new application areas in the near future. Common problems in prediction method usage are discussed. The number of features used for method training or the number of variation types predicted by a tool are not indicators of method performance. Many published gene, protein or disease‐specific benchmark studies suffer from too small dataset rendering the results useless. In the case of binary predictors, equal number of positive and negative cases is beneficial for training, the imbalance has to be corrected for performance assessment. Predictors cannot be better than the data they are based on and used for training and testing. Minor allele frequency (MAF) can help to detect likely benign cases, but the recommended MAF threshold is apparently too high. The fact that many rare variants are disease‐causing or ‐related does not mean that rare variants in general would be harmful. How large a portion of the tested variants a tool can predict (coverage) is not a quality measure.

Conclusion

Methods used for variation interpretation have to be carefully selected. It should be possible to use only one predictor, with proven good performance or a limited number of complementary predictors with state‐of‐the‐art performance. Bear in mind that diseases and pathogenicity have a continuum and variants are not dichotomic i.e. either pathogenic or benign, either.

Prediction of impacts of mutations on protein structure and interactions: SDM, a statistical approach, and mCSM, using machine learning

Next-generation sequencing methods have not only allowed an understanding of genome sequence variation during the evolution of organisms but have also provided invaluable information about genetic variants in inherited disease and the emergence of resistance to drugs in cancers and infectious disease. A challenge is to distinguish mutations that are drivers of disease or drug resistance, from passengers that are neutral or even selectively advantageous to the organism. This requires an understanding of impacts of missense mutations in gene expression and regulation, and on the disruption of protein function by modulating protein stability or disturbing interactions with proteins, nucleic acids, small molecule ligands, and other biological molecules. Experimental approaches to understanding differences between wild-type and mutant proteins are most accurate but are also time-consuming and costly. Computational tools used to predict the impacts of mutations can provide useful information more quickly. Here, we focus on two widely used structure-based approaches, originally developed in the Blundell lab: site-directed mutator (SDM), a statistical approach to analyze amino acid substitutions, and mutation cutoff scanning matrix (mCSM), which uses graph-based signatures to represent the wild-type structural environment and machine learning to predict the effect of mutations on protein stability. Here, we describe DUET that uses machine learning to combine the two approaches. We discuss briefly the development of mCSM for understanding the impacts of mutations on interfaces with other proteins, nucleic acids, and ligands, and we exemplify the wide application of these approaches to understand human genetic disorders and drug resistance mutations relevant to cancer and mycobacterial infections. STATEMENT FOR A BROADER AUDIENCE: Genetic or somatic changes in genes can lead to mutations in human proteins, which give rise to genetic disorders or cancer, or to genes of pathogens leading to drug resistance. Computer software described here, using statistical approaches or machine learning, uses the information from genome sequencing of humans and pathogens, together with experimental or modeled 3D structures of gene products, the proteins, to predict impacts of mutations in genetic disease, cancer and drug resistance.

Computing the Pathogenicity of Wilson's Disease ATP7B Mutations: Implications for Disease Prevalence

Genetic variations in the gene encoding the copper-transport protein ATP7B are the primary cause of Wilson's disease. Controversially, clinical prevalence seems much smaller than the prevalence estimated by genetic screening tools, causing fear that many people are undiagnosed, although early diagnosis and treatment is essential. To address this issue, we benchmarked 16 state-of-the-art computational disease-prediction methods against established data of missense ATP7B mutations. Our results show that the quality of the methods varies widely. We show the importance of optimizing the threshold of the methods used to distinguish pathogenic from nonpathogenic mutations against data of clinically confirmed pathogenic and nonpathogenic mutations. We find that most methods use thresholds that predict too many ATP7B mutations to be pathogenic. Thus, our findings explain the current controversy on Wilson's disease prevalence because meta-analysis and text search methods include many computational estimates that lead to higher disease prevalence than clinically observed. As proteins and diseases differ widely, a one-size-fits-all threshold cannot distinguish pathogenic and nonpathogenic mutations efficiently, as shown here. We also show that amino acid changes with small evolutionary substitution probability, mainly due to amino acid volume, are more associated with the disease, implying a pathological effect on the conformational state of the protein, which could affect copper transport or adenosine triphosphate recognition and hydrolysis. These findings may be a first step toward a more quantitative genotype-phenotype relationship of Wilson's disease.

Predicting Functional Effects of Synonymous Variants: A Systematic Review and Perspectives

Recent advances in high-throughput experimentation have put the exploration of genome sequences at the forefront of precision medicine. In an effort to interpret the sequencing data, numerous computational methods have been developed for evaluating the effects of genome variants. Interestingly, despite the fact that every person has as many synonymous (sSNV) as non-synonymous single nucleotide variants, our ability to predict their effects is limited. The paucity of experimentally tested sSNV effects appears to be the limiting factor in development of such methods. Here, we summarize the details and evaluate the performance of nine existing computational methods capable of predicting sSNV effects. We used a set of observed and artificially generated variants to approximate large scale performance expectations of these tools. We note that the distribution of these variants across amino acid and codon types suggests purifying evolutionary selection retaining generated variants out of the observed set; i.e., we expect the generated set to be enriched for deleterious variants. Closer inspection of the relationship between the observed variant frequencies and the associated prediction scores identifies predictor-specific scoring thresholds of reliable effect predictions. Notably, across all predictors, the variants scoring above these thresholds were significantly more often generated than observed. which confirms our assumption that the generated set is enriched for deleterious variants. Finally, we find that while the methods differ in their ability to identify severe sSNV effects, no predictor appears capable of definitively recognizing subtle effects of such variants on a large scale.

[Rational use of genetic tests in internal medicine : Possibilities and limitations of next generation sequencing diagnostics]

BACKGROUND

New methods of molecular genetic diagnostics enable a more comprehensive genetic analysis of patients.

OBJECTIVES

Rational use and benefits of molecular genetic testing in patients with various internal diseases.

METHOD

Evaluation of topic-related literature, discussion of own experiences, as well as consideration of current guidelines.

RESULTS

New genetic tests, such as next generation sequencing (NGS), improve the diagnosis of hereditary diseases; however, the use of this technology also leads to additional findings, which must be carefully considered.

CONCLUSION

The rational use of genetic tests is a benefit for patients and can significantly influence the prevention and treatment of a disease. The increasing complexity of genetic findings requires interdisciplinary approaches involving human genetics, internal medicine, and other disciplines.

Exploring the use of molecular dynamics in assessing protein variants for phenotypic alterations

With the advent of rapid sequencing technologies, making sense of all the genomic variations that we see among us has been a major challenge. A plethora of algorithms and methods exist that try to address genome interpretation through genotype-phenotype linkage analysis or evaluating the loss of function/stability mutations in protein. Critical Assessment of Genome Interpretation (CAGI) offers an exceptional platform to blind-test all such algorithms and methods to assess their true ability. We take advantage of this opportunity to explore the use of molecular dynamics simulation as a tool to assess alteration of phenotype, loss of protein function, interaction, and stability. The results show that coarse-grained dynamics based protein flexibility analysis on 34 CHEK2 and 1719 CALM1 single mutants perform reasonably well for class-based predictions for phenotype alteration and two-thirds of the predicted scores return a correlation coefficient of 0.6 or more. When all-atom dynamics is used to predict altered stability due to mutations for Frataxin protein (8 cases), the predictions are comparable to the state-of-the-art methods. The competitive performance of our straightforward approach to phenotype interpretation contrasts with heavily trained machine learning approaches, and open new avenues to rationally improve genome interpretation.

Performance of computational methods for the evaluation of pericentriolar material 1 missense variants in CAGI-5

The CAGI-5 pericentriolar material 1 (PCM1) challenge aimed to predict the effect of 38 transgenic human missense mutations in the PCM1 protein implicated in schizophrenia. Participants were provided with 16 benign variants (negative controls), 10 hypomorphic, and 12 loss of function variants. Six groups participated and were asked to predict the probability of effect and standard deviation associated to each mutation. Here, we present the challenge assessment. Prediction performance was evaluated using different measures to conclude in a final ranking which highlights the strengths and weaknesses of each group. The results show a great variety of predictions where some methods performed significantly better than others. Benign variants played an important role as negative controls, highlighting predictors biased to identify disease phenotypes. The best predictor, Bromberg lab, used a neural-network-based method able to discriminate between neutral and non-neutral single nucleotide polymorphisms. The CAGI-5 PCM1 challenge allowed us to evaluate the state of the art techniques for interpreting the effect of novel variants for a difficult target protein.

Benchmarking subcellular localization and variant tolerance predictors on membrane proteins

Background

Membrane proteins constitute up to 30% of the human proteome. These proteins have special properties because the transmembrane segments are embedded into lipid bilayer while extramembranous parts are in different environments. Membrane proteins have several functions and are involved in numerous diseases. A large number of prediction methods have been introduced to predict protein subcellular localization as well as the tolerance or pathogenicity of amino acid substitutions.

Results

We tested the performance of 22 tolerance predictors by collecting information on membrane proteins and variants in them. The analysis indicated that the best tools had similar prediction performance on transmembrane, inside and outside regions of transmembrane proteins and comparable to overall prediction performances for all types of proteins. PON-P2 had the highest performance followed by REVEL, MetaSVM and VEST3. Further, we tested with the high quality dataset also the performance of seven subcellular localization predictors on membrane proteins. We assessed separately the performance for single pass and multi pass membrane proteins. Predictions for multi pass proteins were more reliable than those for single pass proteins.

Conclusions

The predictors for variant effects had better performance than subcellular localization tools. The best tolerance predictors are highly reliable. As there are large differences in the performances of tools, end-users have to be cautious in method selection.

Electronic supplementary material

The online version of this article (10.1186/s12864-019-5865-0) contains supplementary material, which is available to authorized users.

Prediction of disease-associated mutations in the transmembrane regions of proteins with known 3D structure

Being able to assess the phenotypic effects of mutations is a much required capability in precision medicine. However, most of the currently available structure-based methods actually predict stability changes caused by mutations rather than their pathogenic potential. There are also no dedicated methods to predict damaging mutations specifically in transmembrane proteins. In this study we developed and applied a machine-learning approach to discriminate between disease-associated and benign point mutations in the transmembrane regions of proteins with known 3D structure. The method, called BorodaTM (BOosted RegressiOn trees for Disease-Associated mutations in TransMembrane proteins), was trained on sequence-, structure-, and energy-derived descriptors. When compared with the state-of-the-art methods, BorodaTM is superior in classifying point mutations in transmembrane regions. Using BorodaTM we have conducted a large-scale mutation analysis in the transmembrane regions of human proteins with known 3D structures. For each protein we generated structural models for all point mutations by replacing each residue to 19 possible residue types. We classified ~1.8 millions point mutations as benign or diseased-associated and made all predictions available as a Web-server at https://www.iitm.ac.in/bioinfo/MutHTP/boroda.php.

BRCA1- and BRCA2-specific in silico tools for variant interpretation in the CAGI 5 ENIGMA challenge

BRCA1 and BRCA2 (BRCA1/2) germline variants disrupting the DNA protective role of these genes increase the risk of hereditary breast and ovarian cancers. Correct identification of these variants then becomes clinically relevant, because it may increase the survival rates of the carriers. Unfortunately, we are still unable to systematically predict the impact of BRCA1/2 variants. In this article, we present a family of in silico predictors that address this problem, using a gene-specific approach. For each protein, we have developed two tools, aimed at predicting the impact of a variant at two different levels: Functional and clinical. Testing their performance in different datasets shows that specific information compensates the small number of predictive features and the reduced training sets employed to develop our models. When applied to the variants of the BRCA1/2 (ENIGMA) challenge in the fifth Critical Assessment of Genome Interpretation (CAGI 5) we find that these methods, particularly those predicting the functional impact of variants, have a good performance, identifying the large compositional bias towards neutral variants in the CAGI sample. This performance is further improved when incorporating to our prediction protocol estimates of the impact on splicing of the target variant.