Introduction Advancing whole-genome precision medicine requires understanding how gene expression is altered by genetic variants, especially those that are outside of protein-coding regions. We developed a computational technique that scores how strongly genetic variants alter RNA splicing, a critical step in gene expression whose disruption contributes to many diseases, including cancers and neurological disorders. A genome-wide analysis reveals tens of thousands of variants that alter splicing and are enriched with a wide range of known diseases. Our results provide insight into the genetic basis of spinal muscular atrophy, hereditary nonpolyposis colorectal cancer and autism spectrum disorder. Methods We used machine learning to derive a computational model that takes as input DNA sequences and applies general rules to predict splicing in human tissues. Given a test variant, our model computes a score that predicts how much the variant disrupts splicing. The model was derived in such a way that it can be used to study diverse diseases and disorders, and to determine the consequences of common, rare, and even spontaneous variants. Results Our technique is able to accurately classify disease-causing variants and provides insights into the role of aberrant splicing in disease. We scored over 650,000 DNA variants and found that disease-causing variants have higher scores than common variants and even those associated with disease in genome-wide association studies. Our model predicts substantial and unexpected aberrant splicing due to variants within introns and exons, including those far from the splice site. For example, among intronic variants that are more than 30 nucleotides away from a splice site, known disease variants alter splicing nine times more often than common variants; among missense exonic disease variants, those that least impact protein function are over five times more likely to alter splicing than other variants. Autism has been associated with disrupted splicing in brain regions, so we used our method to score variants detected using whole genome sequencing data from individuals with and without autism. Genes with high scoring variants include many that have been previously linked with autism, as well as new genes with known neurodevelopmental phenotypes. Most of the high scoring variants are intronic and cannot be detected by exome analysis techniques. When we score clinical variants in spinal muscular atrophy and colorectal cancer genes, up to 94% of variants found to disrupt splicing using minigene reporters are correctly classified. Discussion In the context of precision medicine, causal support for variants that is independent of existing studies is greatly needed. Our computational model was trained to predict splicing from DNA sequence alone, without using disease annotations or population data. Consequently, its predictions are independent of and complementary to population data, genome-wide association studies (GWAS), expression-based quantitative trait loci (QTL), and functi...

MotivationAlternative splice site selection is inherently competitive and the probability of a given splice site to be used also depends on the strength of neighboring sites. Here, we present a new model named the competitive splice site model (COSSMO), which explicitly accounts for these competitive effects and predicts the percent selected index (PSI) distribution over any number of putative splice sites. We model an alternative splicing event as the choice of a 3′ acceptor site conditional on a fixed upstream 5′ donor site or the choice of a 5′ donor site conditional on a fixed 3′ acceptor site. We build four different architectures that use convolutional layers, communication layers, long short-term memory and residual networks, respectively, to learn relevant motifs from sequence alone. We also construct a new dataset from genome annotations and RNA-Seq read data that we use to train our model.ResultsCOSSMO is able to predict the most frequently used splice site with an accuracy of 70% on unseen test data, and achieve an R2 of 0.6 in modeling the PSI distribution. We visualize the motifs that COSSMO learns from sequence and show that COSSMO recognizes the consensus splice site sequences and many known splicing factors with high specificity.Availability and implementationModel predictions, our training dataset, and code are available from http://cossmo.genes.toronto.edu.Supplementary information Supplementary data are available at Bioinformatics online.

When estimating expression of a transcript or part of a transcript using RNA-seq data, it is commonly assumed that reads are generated uniformly from positions within the transcript. While this assumption is acceptable for long transcript sequences where reads from many positions are averaged, it frequently leads to large errors for short sequences, e.g., less than 100 bp. Analysis of short sequences, such as when studying splice junctions and microRNAs, is increasingly important and necessitates addressing errors in short-sequence expression estimation. Indeed, when we examined RNA-seq data from diverse studies, we found that large errors are introduced by variations in RNA-seq coverage due to sequence content, experimental conditions and sample preparation. 1We developed a technique that we call the positional bootstrap, which quantifies the level of uncertainty in expression induced by nonuniform coverage. Unlike methods that attempt to correct for biases in coverage, but do so by making strong assumptions about the form of those biases, the positional bootstrap can quantify the noise induced by all types of bias, including unknown ones. Results obtained using independently generated RNA-seq datasets show that the positional bootstrap increases the accuracy of estimates of alternative splicing levels, tissue-differential alternative splicing and tissue differential expression, by a factor of up to 10.A Python implementation of the algorithm to quantify splicing levels is freely available from github.com/PSI-Lab/BENTO-Seq.

Alternative splicing selection is inherently competitive and the probability for a given splice site to be used depend strongly on the strength of neighbouring sites. Here we present a new model named Competitive Splicing Site Model (COSSMO) that improves on the start of the art in predicting splice site selection by explicitely modelling these competitive effects. We model an alternative splicing event as the choice of a 3' acceptor site conditional on a fixed upstream 5' donor site, or the choice of a 5' donor site conditional on a fixed 3' acceptor site. Our model is a custom architecture that uses convolutional layers, communication layers, LSTMS, and residual networks, to learn relevant motifs from sequence alone. COSSMO is able to predict the most frequently used splice site with an accuracy of 70% on unseen test data, which compares to only around 35% accuracy for MaxEntScan.