Predicting Splicing from Primary Sequence with Deep Learning.

By Kishore Jaganathan, Sofia Panagiotopoulou, J. McRae, S. Darbandi, David G. Knowles, Yang I. Li, J. Kosmicki, Juan Arbelaez, Wenwu Cui, Grace B. Schwartz and 7 more. Cell, 2019

2,423 citations recorded by the source corpus.

The splicing of pre-mRNAs into mature transcripts is remarkable for its precision, but the mechanisms by which the cellular machinery achieves such specificity are incompletely understood. Here, we describe a deep neural network that accurately predicts splice junctions from an arbitrary pre-mRNA transcript sequence, enabling precise prediction of noncoding genetic variants that cause cryptic splicing. Synonymous and intronic mutations with predicted splice-altering consequence validate at a high rate on RNA-seq and are strongly deleterious in the human population. De novo mutations with predicted splice-altering consequence are significantly enriched in patients with autism and intellectual disability compared to healthy controls and validate against RNA-seq in 21 out of 28 of these patients. We estimate that 9%-11% of pathogenic mutations in patients with rare genetic disorders are caused by this previously underappreciated class of disease variation.

Subfields: Molecular Biology

View Paper: https://www.semanticscholar.org/paper/7fa30c8af436b7df513ef41e2cccd8a2404f37f8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

Citations in the corpus

Artificial intelligence in oncology (2020, background): “56 Abnormal splicing has been frequently observed in cancer, and the splicing machinery is being targeted for therapeutic purposes.”
dbNSFP v4: a comprehensive database of transcript-specific functional predictions and annotations for human nonsynonymous and splice-site SNVs (2020, background): “dicting splice site gain and loss [37], and dbMTS, a comprehensive database for microRNA target site SNVs and their functional predictions [38].”
Artificial intelligence in clinical and genomic diagnostics (2019, background): “SpliceAI, a 32-layer deep neural network, is able to predict both canonical and non-canonical splicing directly from exon–intron junction sequence data [27].”
The Nucleotide Transformer: Building and Evaluating Robust Foundation Models for Human Genomics (2024, methodology): “Notably, our 2.5B 6kb-context model matched the performance of the state-of-the-art SpliceAI-10k [35], which was trained on 15kb input sequences, in addition to other splicing baselines; and outperformed SpliceAI when tested on 6kb input sequences.”
WILDS: A Benchmark of in-the-Wild Distribution Shifts (2020, positive): “…been fairly successful at deciphering functional DNA sequence patterns from learned representations and predicting the consequences of genetic perturbations in contexts in which the models are trained (Avsec et al., 2019; Zhou and Troyanskaya, 2015; Kelley et al., 2016; Jaganathan et al., 2019).”
PRIM1 deficiency causes a distinctive primordial dwarfism syndrome (2020, positive): “Analysis of the c.638+36C>G variant with SpliceAI (Jaganathan et al. 2019), NNSplice (Reese et al. 1997), and MaxEntScan (Yeo and Burge 2004) predicted activation of a cryptic donor site, resulting in the inclusion of 31 nucleotides of intronic sequence following exon 6, expected to disrupt the open”
Ensembl 2021 (2020, methodology): “Other improvements include enriched annotation of potential splicing impact (using scores from SpliceAI (34)), incorporation of phenotype association and variant citation data from DisGeNET (35), and protein annotations from neXtProt (36).”
The UCSC Genome Browser database: 2025 update (2024, neutral): “SpliceAI is an open-source deep learning splicing prediction algorithm that can predict splicing alterations caused by DNA variations ( 7 ).”
Alternative Splicing Regulatory Networks: Functions, Mechanisms, and Evolution. (2019, background): “Such screens will provide a foundation for further in-depth studies and will also facilitate the predictions of phenotypic effects of genetic variation and disease mutations that frequently impact the splicing process (Climente-González et al., 2017; Jaganathan et al., 2019; Ohno et al., 2018; Park ”