CoDP: predicting the impact of unclassified genetic variants in MSH6 by the combination of different properties of the protein

Background

Lynch syndrome is a hereditary cancer predisposition syndrome caused by a mutation in one of the DNA mismatch repair (MMR) genes. About 24% of the mutations identified in Lynch syndrome are missense substitutions and the frequency of missense variants in MSH6 is the highest amongst these MMR genes. Because of this high frequency, the genetic testing was not effectively used in MSH6 so far. We, therefore, developed CoDP (Combination of the Different Properties), a bioinformatics tool to predict the impact of missense variants in MSH6.

Methods

We integrated the prediction results of three methods, namely MAPP, PolyPhen-2 and SIFT. Two other structural properties, namely solvent accessibility and the change in the number of heavy atoms of amino acids in the MSH6 protein, were further combined explicitly. MSH6 germline missense variants classified by their associated clinical and molecular data were used to fit the parameters for the logistic regression model and to assess the prediction. The performance of CoDP was compared with those of other conventional tools, namely MAPP, SIFT, PolyPhen-2 and PON-MMR.

Results

A total of 294 germline missense variants were collected from the variant databases and literature. Of them, 34 variants were available for the parameter training and the prediction performance test. We integrated the prediction results of MAPP, PolyPhen-2 and SIFT, and two other structural properties, namely solvent accessibility and the change in the number of heavy atoms of amino acids in the MSH6 protein, were further combined explicitly. Variants data classified by their associated clinical and molecular data were used to fit the parameters for the logistic regression model and to assess the prediction. The values of the positive predictive value (PPV), the negative predictive value (NPV), sensitivity, specificity and accuracy of the tools were compared on the whole data set. PPV of CoDP was 93.3% (14/15), NPV was 94.7% (18/19), specificity was 94.7% (18/19), sensitivity was 93.3% (14/15) and accuracy was 94.1% (32/34). Area under the curve of CoDP was 0.954, that of MAPP for MSH6 was 0.919, of SIFT was 0.864 and of PolyPhen-2 HumVar was 0.819. The power to distinguish between pathogenic and non-pathogenic variants of these methods was tested by Wilcoxon rank sum test (p < 8.9 × 10^-6 for CoDP, p < 3.3 × 10^-5 for MAPP, p < 3.1 × 10^-4 for SIFT and p < 1.2 × 10^-3 for PolyPhen-2 HumVar), and CoDP was shown to outperform other conventional methods.

Conclusion

In this paper, we provide a human curated data set for MSH6 missense variants, and CoDP, the prediction tool, which achieved better accuracy for predicting the impact of missense variants in MSH6 than any other known tools. CoDP is available at http://cib.cf.ocha.ac.jp/CoDP/.     

Characterization of intrinsically disordered regions in proteins informed by human genetic diversity

All proteomes contain both proteins and polypeptide segments that don’t form a defined three-dimensional structure yet are biologically active—called intrinsically disordered proteins and regions (IDPs and IDRs). Most of these IDPs/IDRs lack useful functional annotation limiting our understanding of their importance for organism fitness. Here we characterized IDRs using protein sequence annotations of functional sites and regions available in the UniProt knowledgebase (“UniProt features”: active site, ligand-binding pocket, regions mediating protein-protein interactions, etc.). By measuring the statistical enrichment of twenty-five UniProt features in 981 IDRs of 561 human proteins, we identified eight features that are commonly located in IDRs. We then collected the genetic variant data from the general population and patient-based databases and evaluated the prevalence of population and pathogenic variations in IDPs/IDRs. We observed that some IDRs tolerate 2 to 12-times more single amino acid-substituting missense mutations than synonymous changes in the general population. However, we also found that 37% of all germline pathogenic mutations are located in disordered regions of 96 proteins. Based on the observed-to-expected frequency of mutations, we categorized 34 IDRs in 20 proteins (DDX3X, KIT, RB1, etc.) as intolerant to mutation. Finally, using statistical analysis and a machine learning approach, we demonstrate that mutation-intolerant IDRs carry a distinct signature of functional features. Our study presents a novel approach to assign functional importance to IDRs by leveraging the wealth of available genetic data, which will aid in a deeper understating of the role of IDRs in biological processes and disease mechanisms.     

PDBsum extras: SARS‐CoV‐2 and AlphaFold models

The PDBsum web server provides structural analyses of the entries in the Protein Data Bank (PDB). Two recent additions are described here. The first is the detailed analysis of the SARS‐CoV‐2 virus protein structures in the PDB. These include the variants of concern, which are shown both on the sequences and 3D structures of the proteins. The second addition is the inclusion of the available AlphaFold models for human proteins. The pages allow a search of the protein against existing structures in the PDB via the Sequence Annotated by Structure (SAS) server, so one can easily compare the predicted model against experimentally determined structures. The server is freely accessible to all at http://www.ebi.ac.uk/pdbsum.     

In silico prediction of HIV-1-host molecular interactions and their directionality

Human immunodeficiency virus type 1 (HIV-1) continues to be a major cause of disease and premature death. As with all viruses, HIV-1 exploits a host cell to replicate. Improving our understanding of the molecular interactions between virus and human host proteins is crucial for a mechanistic understanding of virus biology, infection and host antiviral activities. This knowledge will potentially permit the identification of host molecules for targeting by drugs with antiviral properties. Here, we propose a data-driven approach for the analysis and prediction of the HIV-1 interacting proteins (VIPs) with a focus on the directionality of the interaction: host-dependency versus antiviral factors. Using support vector machine learning models and features encompassing genetic, proteomic and network properties, our results reveal some significant differences between the VIPs and non-HIV-1 interacting human proteins (non-VIPs). As assessed by comparison with the HIV-1 infection pathway data in the Reactome database (sensitivity > 90%, threshold = 0.5), we demonstrate these models have good generalization properties. We find that the ‘direction’ of the HIV-1-host molecular interactions is also predictable due to different characteristics of ‘forward’/pro-viral versus ‘backward’/pro-host proteins. Additionally, we infer the previously unknown direction of the interactions between HIV-1 and 1351 human host proteins. A web server for performing predictions is available at http://hivpre.cvr.gla.ac.uk/.     

Prediction of Spontaneous Protein Deamidation from Sequence-Derived Secondary Structure and Intrinsic Disorder

Asparagine residues in proteins undergo spontaneous deamidation, a post-translational modification that may act as a molecular clock for the regulation of protein function and turnover. Asparagine deamidation is modulated by protein local sequence, secondary structure and hydrogen bonding. We present NGOME, an algorithm able to predict non-enzymatic deamidation of internal asparagine residues in proteins in the absence of structural data, using sequence-based predictions of secondary structure and intrinsic disorder. Compared to previous algorithms, NGOME does not require three-dimensional structures yet yields better predictions than available sequence-only methods. Four case studies of specific proteins show how NGOME may help the user identify deamidation-prone asparagine residues, often related to protein gain of function, protein degradation or protein misfolding in pathological processes. A fifth case study applies NGOME at a proteomic scale and unveils a correlation between asparagine deamidation and protein degradation in yeast. NGOME is freely available as a webserver at the National EMBnet node Argentina, URL: http://www.embnet.qb.fcen.uba.ar/ in the subpage “Protein and nucleic acid structure and sequence analysis”.     

Building a Better Fragment Library for De Novo Protein Structure Prediction

Fragment-based approaches are the current standard for de novo protein structure prediction. These approaches rely on accurate and reliable fragment libraries to generate good structural models. In this work, we describe a novel method for structure fragment library generation and its application in fragment-based de novo protein structure prediction. The importance of correct testing procedures in assessing the quality of fragment libraries is demonstrated. In particular, the exclusion of homologs to the target from the libraries to correctly simulate a de novo protein structure prediction scenario, something which surprisingly is not always done. We demonstrate that fragments presenting different predominant predicted secondary structures should be treated differently during the fragment library generation step and that exhaustive and random search strategies should both be used. This information was used to develop a novel method, Flib. On a validation set of 41 structurally diverse proteins, Flib libraries presents both a higher precision and coverage than two of the state-of-the-art methods, NNMake and HHFrag. Flib also achieves better precision and coverage on the set of 275 protein domains used in the two previous experiments of the the Critical Assessment of Structure Prediction (CASP9 and CASP10). We compared Flib libraries against NNMake libraries in a structure prediction context. Of the 13 cases in which a correct answer was generated, Flib models were more accurate than NNMake models for 10. “Flib is available for download at: http://www.stats.ox.ac.uk/research/proteins/resources”.     

Evaluation and Cross-Comparison of Lexical Entities of Biological Interest (LexEBI)

Motivation

Biomedical entities, their identifiers and names, are essential in the representation of biomedical facts and knowledge. In the same way, the complete set of biomedical and chemical terms, i.e. the biomedical “term space” (the “Lexeome”), forms a key resource to achieve the full integration of the scientific literature with biomedical data resources: any identified named entity can immediately be normalized to the correct database entry. This goal does not only require that we are aware of all existing terms, but would also profit from knowing all their senses and their semantic interpretation (ambiguities, nestedness).

Result

This study compiles a resource for lexical terms of biomedical interest in a standard format (called “LexEBI”), determines the overall number of terms, their reuse in different resources and the nestedness of terms. LexEBI comprises references for protein and gene entries and their term variants and chemical entities amongst other terms. In addition, disease terms have been identified from Medline and PubmedCentral and added to LexEBI. Our analysis demonstrates that the baseforms of terms from the different semantic types show only little polysemous use. Nonetheless, the term variants of protein and gene names (PGNs) frequently contain species mentions, which should have been avoided according to protein annotation guidelines. Furthermore, the protein and gene entities as well as the chemical entities, both do comprise enzymes leading to hierarchical polysemy, and a large portion of PGNs make reference to a chemical entity. Altogether, according to our analysis based on the Medline distribution, 401,869 unique PGNs in the documents contain a reference to 25,022 chemical entities, 3,125 disease terms or 1,576 species mentions.

Conclusion

LexEBI delivers the complete biomedical and chemical Lexeome in a standardized representation (http://www.ebi.ac.uk/Rebholz-srv/LexEBI/). The resource provides the disease terms as open source content, and fully interlinks terms across resources.     

Can Predicted Protein 3D Structures Provide Reliable Insights into whether Missense Variants Are Disease Associated?

Knowledge of protein structure can be used to predict the phenotypic consequence of a missense variant. Since structural coverage of the human proteome can be roughly tripled to over 50% of the residues if homology-predicted structures are included in addition to experimentally determined coordinates, it is important to assess the reliability of using predicted models when analyzing missense variants. Accordingly, we assess whether a missense variant is structurally damaging by using experimental and predicted structures. We considered 606 experimental structures and show that 40% of the 1965 disease-associated missense variants analyzed have a structurally damaging change in the mutant structure. Only 11% of the 2134 neutral variants are structurally damaging. Importantly, similar results are obtained when 1052 structures predicted using Phyre2 algorithm were used, even when the model shares low (< 40%) sequence identity to the template. Thus, structure-based analysis of the effects of missense variants can be effectively applied to homology models. Our in-house pipeline, Missense3D, for structurally assessing missense variants was made available at http://www.sbg.bio.ic.ac.uk/~missense3d     

VarSite: Disease variants and protein structure

VarSite is a web server mapping known disease‐associated variants from UniProt and ClinVar, together with natural variants from gnomAD, onto protein 3D structures in the Protein Data Bank. The analyses are primarily image‐based and provide both an overview for each human protein, as well as a report for any specific variant of interest. The information can be useful in assessing whether a given variant might be pathogenic or benign. The structural annotations for each position in the protein include protein secondary structure, interactions with ligand, metal, DNA/RNA, or other protein, and various measures of a given variant's possible impact on the protein's function. The 3D locations of the disease‐associated variants can be viewed interactively via the 3dmol.js JavaScript viewer, as well as in RasMol and PyMOL. Users can search for specific variants, or sets of variants, by providing the DNA coordinates of the base change(s) of interest. Additionally, various agglomerative analyses are given, such as the mapping of disease and natural variants onto specific Pfam or CATH domains. The server is freely accessible to all at: https://www.ebi.ac.uk/thornton-srv/databases/VarSite .     

BRCA1 and BRCA2 missense variants of high and low clinical significance influence lymphoblastoid cell line post-irradiation gene expression

The functional consequences of missense variants in disease genes are difficult to predict. We assessed if gene expression profiles could distinguish between BRCA1 or BRCA2 pathogenic truncating and missense mutation carriers and familial breast cancer cases whose disease was not attributable to BRCA1 or BRCA2 mutations (BRCAX cases). 72 cell lines from affected women in high-risk breast ovarian families were assayed after exposure to ionising irradiation, including 23 BRCA1 carriers, 22 BRCA2 carriers, and 27 BRCAX individuals. A subset of 10 BRCAX individuals carried rare BRCA1/2 sequence variants considered to be of low clinical significance (LCS). BRCA1 and BRCA2 mutation carriers had similar expression profiles, with some subclustering of missense mutation carriers. The majority of BRCAX individuals formed a distinct cluster, but BRCAX individuals with LCS variants had expression profiles similar to BRCA1/2 mutation carriers. Gaussian Process Classifier predicted BRCA1, BRCA2 and BRCAX status, with a maximum of 62% accuracy, and prediction accuracy decreased with inclusion of BRCAX samples carrying an LCS variant, and inclusion of pathogenic missense carriers. Similarly, prediction of mutation status with gene lists derived using Support Vector Machines was good for BRCAX samples without an LCS variant (82–94%), poor for BRCAX with an LCS (40–50%), and improved for pathogenic BRCA1/2 mutation carriers when the gene list used for prediction was appropriate to mutation effect being tested (71–100%). This study indicates that mutation effect, and presence of rare variants possibly associated with a low risk of cancer, must be considered in the development of array-based assays of variant pathogenicity.     

Predicting Transmembrane Helix Packing Arrangements using Residue Contacts and a Force-Directed Algorithm

Alpha-helical transmembrane proteins constitute roughly 30% of a typical genome and are involved in a wide variety of important biological processes including cell signalling, transport of membrane-impermeable molecules and cell recognition. Despite significant efforts to predict transmembrane protein topology, comparatively little attention has been directed toward developing a method to pack the helices together. Here, we present a novel approach to predict lipid exposure, residue contacts, helix-helix interactions and finally the optimal helical packing arrangement of transmembrane proteins. Using molecular dynamics data, we have trained and cross-validated a support vector machine (SVM) classifier to predict per residue lipid exposure with 69% accuracy. This information is combined with additional features to train a second SVM to predict residue contacts which are then used to determine helix-helix interaction with up to 65% accuracy under stringent cross-validation on a non-redundant test set. Our method is also able to discriminate native from decoy helical packing arrangements with up to 70% accuracy. Finally, we employ a force-directed algorithm to construct the optimal helical packing arrangement which demonstrates success for proteins containing up to 13 transmembrane helices. This software is freely available as source code from http://bioinf.cs.ucl.ac.uk/memsat/mempack/.     

Pathogenic BCL11A variants provide insights into the mechanisms of human fetal hemoglobin silencing

Increased production of fetal hemoglobin (HbF) can ameliorate the severity of sickle cell disease and β-thalassemia. BCL11A has been identified as a key regulator of HbF silencing, although its precise mechanisms of action remain incompletely understood. Recent studies have identified pathogenic mutations that cause heterozygous loss-of-function of BCL11A and result in a distinct neurodevelopmental disorder that is characterized by persistent HbF expression. While the majority of cases have deletions or null mutations causing haploinsufficiency of BCL11A, several missense variants have also been identified. Here, we perform functional studies on these variants to uncover specific liabilities for BCL11A’s function in HbF silencing. We find several mutations in an N-terminal C2HC zinc finger that increase proteasomal degradation of BCL11A. We also identify a distinct C-terminal missense variant in the fifth zinc finger domain that we demonstrate causes loss-of-function through disruption of DNA binding. Our analysis of missense variants causing loss-of-function in vivo illuminates mechanisms by which BCL11A silences HbF and also suggests potential therapeutic avenues for HbF induction to treat sickle cell disease and β-thalassemia.     

PremPS: Predicting the impact of missense mutations on protein stability

Computational methods that predict protein stability changes induced by missense mutations have made a lot of progress over the past decades. Most of the available methods however have very limited accuracy in predicting stabilizing mutations because existing experimental sets are dominated by mutations reducing protein stability. Moreover, few approaches could consistently perform well across different test cases. To address these issues, we developed a new computational method PremPS to more accurately evaluate the effects of missense mutations on protein stability. The PremPS method is composed of only ten evolutionary- and structure-based features and parameterized on a balanced dataset with an equal number of stabilizing and destabilizing mutations. A comprehensive comparison of the predictive performance of PremPS with other available methods on nine benchmark datasets confirms that our approach consistently outperforms other methods and shows considerable improvement in estimating the impacts of stabilizing mutations. A protein could have multiple structures available, and if another structure of the same protein is used, the predicted change in stability for structure-based methods might be different. Thus, we further estimated the impact of using different structures on prediction accuracy, and demonstrate that our method performs well across different types of structures except for low-resolution structures and models built based on templates with low sequence identity. PremPS can be used for finding functionally important variants, revealing the molecular mechanisms of functional influences and protein design. PremPS is freely available at https://lilab.jysw.suda.edu.cn/research/PremPS/, which allows to do large-scale mutational scanning and takes about four minutes to perform calculations for a single mutation per protein with ~ 300 residues and requires ~ 0.4 seconds for each additional mutation.     

The eSNV-detect: a computational system to identify expressed single nucleotide variants from transcriptome sequencing data

Rapid development of next generation sequencing technology has enabled the identification of genomic alterations from short sequencing reads. There are a number of software pipelines available for calling single nucleotide variants from genomic DNA but, no comprehensive pipelines to identify, annotate and prioritize expressed SNVs (eSNVs) from non-directional paired-end RNA-Seq data. We have developed the eSNV-Detect, a novel computational system, which utilizes data from multiple aligners to call, even at low read depths, and rank variants from RNA-Seq. Multi-platform comparisons with the eSNV-Detect variant candidates were performed. The method was first applied to RNA-Seq from a lymphoblastoid cell-line, achieving 99.7% precision and 91.0% sensitivity in the expressed SNPs for the matching HumanOmni2.5 BeadChip data. Comparison of RNA-Seq eSNV candidates from 25 ER+ breast tumors from The Cancer Genome Atlas (TCGA) project with whole exome coding data showed 90.6–96.8% precision and 91.6–95.7% sensitivity. Contrasting single-cell mRNA-Seq variants with matching traditional multicellular RNA-Seq data for the MD-MB231 breast cancer cell-line delineated variant heterogeneity among the single-cells. Further, Sanger sequencing validation was performed for an ER+ breast tumor with paired normal adjacent tissue validating 29 out of 31 candidate eSNVs. The source code and user manuals of the eSNV-Detect pipeline for Sun Grid Engine and virtual machine are available at http://bioinformaticstools.mayo.edu/research/esnv-detect/.     

Ensuring the quality and specificity of preregistrations

Researchers face many, often seemingly arbitrary, choices in formulating hypotheses, designing protocols, collecting data, analyzing data, and reporting results. Opportunistic use of “researcher degrees of freedom” aimed at obtaining statistical significance increases the likelihood of obtaining and publishing false-positive results and overestimated effect sizes. Preregistration is a mechanism for reducing such degrees of freedom by specifying designs and analysis plans before observing the research outcomes. The effectiveness of preregistration may depend, in part, on whether the process facilitates sufficiently specific articulation of such plans. In this preregistered study, we compared 2 formats of preregistration available on the OSF: Standard Pre-Data Collection Registration and Prereg Challenge Registration (now called “OSF Preregistration,” http://osf.io/prereg/). The Prereg Challenge format was a “structured” workflow with detailed instructions and an independent review to confirm completeness; the “Standard” format was “unstructured” with minimal direct guidance to give researchers flexibility for what to prespecify. Results of comparing random samples of 53 preregistrations from each format indicate that the “structured” format restricted the opportunistic use of researcher degrees of freedom better (Cliff’s Delta = 0.49) than the “unstructured” format, but neither eliminated all researcher degrees of freedom. We also observed very low concordance among coders about the number of hypotheses (14%), indicating that they are often not clearly stated. We conclude that effective preregistration is challenging, and registration formats that provide effective guidance may improve the quality of research.

Researchers face many, often seemingly arbitrary choices in formulating hypotheses, designing protocols, collecting data, analyzing data, and reporting results. A study of two formats of preregistration available on the OSF reveals that the opportunistic use of researcher degrees of freedom aimed at obtaining statistical significance is restricted by using more extensive preregistration guidelines; however, these guidelines should be further improved.