Investigating DNA‐, RNA‐, and protein‐based features as a means to discriminate pathogenic synonymous variants. Issue 10 (10th July 2017)

• Record Type: Journal Article
• Authors: Livingstone, Mark; Folkman, Lukas; Yang, Yuedong; Zhang, Ping; Mort, Matthew; Cooper, David N.; Liu, Yunlong; Stantic, Bela; Zhou, Yaoqi
• Is Part Of: Human mutation. Volume 38:Issue 10(2017)
• Page Start: 1336
• Page End: 1347
• Publication Date: 2017-07-10
• Subjects: bioinformatics -- machine learning -- same‐sense variant -- silent mutation -- synonymous SNV; Human chromosome abnormalities -- Periodicals; Mutation (Biology) -- Periodicals; 616.04205
• DOI: 10.1002/humu.23283
• Languages: English
• ISSNs: 1059-7794

Abstract

Abstract : The article presents a machine learning model, termed DDIG‐SN, as a means to discriminate disease‐causing synonymous variants. For training DDIG‐SN, we evaluated a range of DNA‐, RNA‐, and protein‐level features. We were able to show that the disease‐causing effects in the immediate proximity to exon‐intron junctions (1–3 bp) are driven by the loss of splicing motif strength, whereas the gain of splicing motif strength is the primary cause in regions further away from the splice site (4–69 bp). Abstract: Synonymous single‐nucleotide variants (SNVs), although they do not alter the encoded protein sequences, have been implicated in many genetic diseases. Experimental studies indicate that synonymous SNVs can lead to changes in the secondary and tertiary structures of DNA and RNA, thereby affecting translational efficiency, cotranslational protein folding as well as the binding of DNA‐/RNA‐binding proteins. However, the importance of these various features in disease phenotypes is not clearly understood. Here, we have built a support vector machine (SVM) model (termed DDIG‐SN) as a means to discriminate disease‐causing synonymous variants. The model was trained and evaluated on nearly 900 disease‐causing variants. The method achieves robust performance with the area under the receiver operating characteristic curve of 0.84 and 0.85 for protein‐stratified 10‐fold cross‐validation and independent testing, respectively. We were able to show that the disease‐causing effects in the immediate proximity to exon–intron junctions (1–3 bp) are driven by the loss of splicing motif strength, whereas the gain of splicing motif strength is the primary cause in regions further away from the splice site (4–69 bp). The method is available as a part of the DDIG server athttp://sparks-lab.org/ddig .