GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction.
Ontology highlight
ABSTRACT:
SUBMITTER: Benegas G
PROVIDER: S-EPMC10592768 | biostudies-literature | 2023 Oct
REPOSITORIES: biostudies-literature
Publications
GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction.
Benegas Gonzalo G Albors Carlos C Aw Alan J AJ Ye Chengzhong C Song Yun S YS
bioRxiv : the preprint server for biology 20240406
Whereas protein language models have demonstrated remarkable efficacy in predicting the effects of missense variants, DNA counterparts have not yet achieved a similar competitive edge for genome-wide variant effect predictions, especially in complex genomes such as that of humans. To address this challenge, we here introduce GPN-MSA, a novel framework for DNA language models that leverages whole-genome sequence alignments across multiple species and takes only a few hours to train. Across severa ...[more]
Similar Datasets
Project description:The expanding catalog of genome-wide association studies (GWAS) provides biological insights across a variety of species, but identifying the causal variants behind these associations remains a significant challenge. Experimental validation is both labor-intensive and costly, highlighting the need for accurate, scalable computational methods to predict the effects of genetic variants across the entire genome. Inspired by recent progress in natural language processing, unsupervised pretraining on large protein sequence databases has proven successful in extracting complex information related to proteins. These models showcase their ability to learn variant effects in coding regions using an unsupervised approach. Expanding on this idea, we here introduce the Genomic Pre-trained Network (GPN), a model designed to learn genome-wide variant effects through unsupervised pretraining on genomic DNA sequences. Our model also successfully learns gene structure and DNA motifs without any supervision. To demonstrate its utility, we train GPN on unaligned reference genomes of Arabidopsis thaliana and seven related species within the Brassicales order and evaluate its ability to predict the functional impact of genetic variants in A. thaliana by utilizing allele frequencies from the 1001 Genomes Project and a comprehensive database of GWAS. Notably, GPN outperforms predictors based on popular conservation scores such as phyloP and phastCons. Our predictions for A. thaliana can be visualized as sequence logos in the UCSC Genome Browser (https://genome.ucsc.edu/s/gbenegas/gpn-arabidopsis). We provide code (https://github.com/songlab-cal/gpn) to train GPN for any given species using its DNA sequence alone, enabling unsupervised prediction of variant effects across the entire genome.
| S-EPMC10622914 | biostudies-literature
Project description:Predicting the effects of coding variants is a major challenge. While recent deep-learning models have improved variant effect prediction accuracy, they cannot analyze all coding variants due to dependency on close homologs or software limitations. Here we developed a workflow using ESM1b, a 650-million-parameter protein language model, to predict all ~450 million possible missense variant effects in the human genome, and made all predictions available on a web portal. ESM1b outperformed existing methods in classifying ~150,000 ClinVar/HGMD missense variants as pathogenic or benign and predicting measurements across 28 deep mutational scan datasets. We further annotated ~2 million variants as damaging only in specific protein isoforms, demonstrating the importance of considering all isoforms when predicting variant effects. Our approach also generalizes to more complex coding variants such as in-frame indels and stop-gains. Together, these results establish protein language models as an effective, accurate and general approach to predicting variant effects.
| S-EPMC10484790 | biostudies-literature
Project description:BackgroundSplicing of genomic exons into mRNAs is a critical prerequisite for the accurate synthesis of human proteins. Genetic variants impacting splicing underlie a substantial proportion of genetic disease, but are challenging to identify beyond those occurring at donor and acceptor dinucleotides. To address this, various methods aim to predict variant effects on splicing. Recently, deep neural networks (DNNs) have been shown to achieve better results in predicting splice variants than other strategies.MethodsIt has been unclear how best to integrate such process-specific scores into genome-wide variant effect predictors. Here, we use a recently published experimental data set to compare several machine learning methods that score variant effects on splicing. We integrate the best of those approaches into general variant effect prediction models and observe the effect on classification of known pathogenic variants.ResultsWe integrate two specialized splicing scores into CADD (Combined Annotation Dependent Depletion; cadd.gs.washington.edu ), a widely used tool for genome-wide variant effect prediction that we previously developed to weight and integrate diverse collections of genomic annotations. With this new model, CADD-Splice, we show that inclusion of splicing DNN effect scores substantially improves predictions across multiple variant categories, without compromising overall performance.ConclusionsWhile splice effect scores show superior performance on splice variants, specialized predictors cannot compete with other variant scores in general variant interpretation, as the latter account for nonsense and missense effects that do not alter splicing. Although only shown here for splice scores, we believe that the applied approach will generalize to other specific molecular processes, providing a path for the further improvement of genome-wide variant effect prediction.
| S-EPMC7901104 | biostudies-literature
Project description:Genetically modified genomes are often used today in many areas of fundamental and applied research. In many studies, coding or noncoding regions are modified in order to change protein sequences or gene expression levels. Modifying one or several nucleotides in a genome can also lead to unexpected changes in the epigenetic regulation of genes. When designing a synthetic genome with many mutations, it would thus be very informative to be able to predict the effect of these mutations on chromatin. We develop here a deep learning approach that quantifies the effect of every possible single mutation on nucleosome positions on the full Saccharomyces cerevisiae genome. This type of annotation track can be used when designing a modified S. cerevisiae genome. We further highlight how this track can provide new insights on the sequence-dependent mechanisms that drive nucleosomes' positions in vivo.
| S-EPMC7849406 | biostudies-literature
Project description:DNA methylation in repetitive elements (RE) suppresses their mobility and maintains genomic stability, and decreases in it are frequently observed in tumor and/or surrogate tissues. Averaging methylation across RE in genome is widely used to quantify global methylation. However, methylation may vary in specific RE and play diverse roles in disease development, thus averaging methylation across RE may lose significant biological information. The ambiguous mapping of short reads by and high cost of current bisulfite sequencing platforms make them impractical for quantifying locus-specific RE methylation. Although microarray-based approaches (particularly Illumina's Infinium methylation arrays) provide cost-effective and robust genome-wide methylation quantification, the number of interrogated CpGs in RE remains limited. We report a random forest-based algorithm (and corresponding R package, REMP) that can accurately predict genome-wide locus-specific RE methylation based on Infinium array profiling data. We validated its prediction performance using alternative sequencing and microarray data. Testing its clinical utility with The Cancer Genome Atlas data demonstrated that our algorithm offers more comprehensively extended locus-specific RE methylation information that can be readily applied to large human studies in a cost-effective manner. Our work has the potential to improve our understanding of the role of global methylation in human diseases, especially cancer.
| S-EPMC5587781 | biostudies-literature
Project description:BackgroundDNA methylation (DNAm) has important regulatory roles in many biological processes and diseases. It is the only epigenetic mark with a clear mechanism of mitotic inheritance and the only one easily available on a genome scale. Aberrant cytosine-phosphate-guanine (CpG) methylation has been discussed in the context of disease aetiology, especially cancer. CpG hypermethylation of promoter regions is often associated with silencing of tumour suppressor genes and hypomethylation with activation of oncogenes.Supervised principal component analysis (SPCA) is a popular machine learning method. However, in a recent application to phenotype prediction from DNAm data SPCA was inferior to the specific method EVORA.ResultsWe present Model-Selection-SPCA (MS-SPCA), an enhanced version of SPCA. MS-SPCA applies several models that perform well in the training data to the test data and selects the very best models for final prediction based on parameters of the test data.We have applied MS-SPCA for phenotype prediction from genome-wide DNAm data. CpGs used for prediction are selected based on the quantification of three features of their methylation (average methylation difference, methylation variation difference and methylation-age-correlation). We analysed four independent case-control datasets that correspond to different stages of cervical cancer: (i) cases currently cytologically normal, but will later develop neoplastic transformations, (ii, iii) cases showing neoplastic transformations and (iv) cases with confirmed cancer. The first dataset was split into several smaller case-control datasets (samples either Human Papilloma Virus (HPV) positive or negative). We demonstrate that cytology normal HPV+ and HPV- samples contain DNAm patterns which are associated with later neoplastic transformations. We present evidence that DNAm patterns exist in cytology normal HPV- samples that (i) predispose to neoplastic transformations after HPV infection and (ii) predispose to HPV infection itself. MS-SPCA performs significantly better than EVORA.ConclusionsMS-SPCA can be applied to many classification problems. Additional improvements could include usage of more than one principal component (PC), with automatic selection of the optimal number of PCs. We expect that MS-SPCA will be useful for analysing recent larger DNAm data to predict future neoplastic transformations.
| S-EPMC4073816 | biostudies-literature
Project description:Reading and language abilities are heritable traits that are likely to share some genetic influences with each other. To identify pleiotropic genetic variants affecting these traits, we first performed a genome-wide association scan (GWAS) meta-analysis using three richly characterized datasets comprising individuals with histories of reading or language problems, and their siblings. GWAS was performed in a total of 1862 participants using the first principal component computed from several quantitative measures of reading- and language-related abilities, both before and after adjustment for performance IQ. We identified novel suggestive associations at the SNPs rs59197085 and rs5995177 (uncorrected P ≈ 10(-7) for each SNP), located respectively at the CCDC136/FLNC and RBFOX2 genes. Each of these SNPs then showed evidence for effects across multiple reading and language traits in univariate association testing against the individual traits. FLNC encodes a structural protein involved in cytoskeleton remodelling, while RBFOX2 is an important regulator of alternative splicing in neurons. The CCDC136/FLNC locus showed association with a comparable reading/language measure in an independent sample of 6434 participants from the general population, although involving distinct alleles of the associated SNP. Our datasets will form an important part of on-going international efforts to identify genes contributing to reading and language skills.
| S-EPMC4165772 | biostudies-literature
Project description:Long noncoding RNAs (lncRNAs) play a key role in many cellular processes including chromatin regulation. To modify chromatin, lncRNAs often interact with DNA in a sequence-specific manner forming RNA:DNA triple helices. Computational tools for triple helix search do not always provide genome-wide predictions of sufficient quality. Here, we used four human lncRNAs (MEG3, DACOR1, TERC and HOTAIR) and their experimentally determined binding regions for evaluating triplex parameters that provide the highest prediction accuracy. Additionally, we combined triplex prediction with the lncRNA secondary structure and demonstrated that considering only single-stranded fragments of lncRNA can further improve DNA-RNA triplexes prediction.
| S-EPMC7037363 | biostudies-literature
Project description:DNA methylation plays a significant role in transcriptional regulation by repressing activity. Change of the DNA methylation level is an important factor affecting the expression of target genes and downstream phenotypes. Because current experimental technologies can only assay a small proportion of CpG sites in the human genome, it is urgent to develop reliable computational models for predicting genome-wide DNA methylation. Here, we proposed a novel algorithm that accurately extracted sequence complexity features (seven features) and developed a support-vector-machine-based prediction model with integration of the reported DNA composition features (trinucleotide frequency and GC content, 65 features) by utilizing the methylation profiles of embryonic stem cells in human. The prediction results from 22 human chromosomes with size-varied windows showed that the 600-bp window achieved the best average accuracy of 94.7%. Moreover, comparisons with two existing methods further showed the superiority of our model, and cross-species predictions on mouse data also demonstrated that our model has certain generalization ability. Finally, a statistical test of the experimental data and the predicted data on functional regions annotated by ChromHMM found that six out of 10 regions were consistent, which implies reliable prediction of unassayed CpG sites. Accordingly, we believe that our novel model will be useful and reliable in predicting DNA methylation.
| S-EPMC5343954 | biostudies-literature
Project description:The RNA 3D Structure-to-Multiple Sequence Alignment Server (R3D-2-MSA) is a new web service that seamlessly links RNA three-dimensional (3D) structures to high-quality RNA multiple sequence alignments (MSAs) from diverse biological sources. In this first release, R3D-2-MSA provides manual and programmatic access to curated, representative ribosomal RNA sequence alignments from bacterial, archaeal, eukaryal and organellar ribosomes, using nucleotide numbers from representative atomic-resolution 3D structures. A web-based front end is available for manual entry and an Application Program Interface for programmatic access. Users can specify up to five ranges of nucleotides and 50 nucleotide positions per range. The R3D-2-MSA server maps these ranges to the appropriate columns of the corresponding MSA and returns the contents of the columns, either for display in a web browser or in JSON format for subsequent programmatic use. The browser output page provides a 3D interactive display of the query, a full list of sequence variants with taxonomic information and a statistical summary of distinct sequence variants found. The output can be filtered and sorted in the browser. Previous user queries can be viewed at any time by resubmitting the output URL, which encodes the search and re-generates the results. The service is freely available with no login requirement at http://rna.bgsu.edu/r3d-2-msa.
| S-EPMC4489222 | biostudies-literature