Abstract
Many problems of modern genetics and functional genomics require the assessment of functional effects of sequence variants, including gene expression changes. Machine learning is considered to be a promising approach for solving this task, but its practical applications remain a challenge due to the insufficient volume and diversity of training data. A promising source of valuable data is a saturation mutagenesis massively parallel reporter assay, which quantitatively measures changes in transcription activity caused by sequence variants. Here, we explore the computational predictions of the effects of individual single-nucleotide variants on gene transcription measured in the massively parallel reporter assays, based on the data from the recent “Regulation Saturation” Critical Assessment of Genome Interpretation challenge. We show that the estimated prediction quality strongly depends on the structure of the training and validation data. Particularly, training on the sequence segments located next to the validation data results in the “information leakage” caused by the local context. This information leakage allows reproducing the prediction quality of the best CAGI challenge submissions with a fairly simple machine learning approach, and even obtaining notably better-than-random predictions using irrelevant genomic regions. Validation scenarios preventing such information leakage dramatically reduce the measured prediction quality. The performance at independent regulatory regions entirely excluded from the training set appears to be much lower than needed for practical applications, and even the performance estimation will become reliable only in the future with richer data from multiple reporters. The source code and data are available at https://bitbucket.org/autosomeru_cagi2018/cagi2018_regsat and https://genomeinterpretation.org/content/expression-variants.
Highlights
Recent progress in medical genetics has drawn attention to sequence variants in the regulatory regions that can alter transcription factor binding (Deplancke et al, 2016), affect cell identity (Liu et al, 2017), and bring about disorders like cancer (Killela et al, 2013) and schizophrenia (Fabbri and Serretti, 2017)
The CAGI “Regulation Saturation” challenge data (Shigaki et al, 2019) included expression changes observed for more than 17 thousand induced single-nucleotide variants (SNVs) within regulatory regions using reporters constructed from 5 human enhancers (IRF4, IRF6, MYC, SORT1, ZFAND3), and 9 promoters (F9, GP1BB, HBB, HBG, HNF4A, LDLR, MSMB, PKLR, TERT), each tested in a particular cell type (TERT was tested in two cell types)
massive parallel reporter assays (MPRA) and machine learning are two recent technologies, the power of which is yet to be harnessed for the progress of genetic studies, in regulatory genomics
Summary
Recent progress in medical genetics has drawn attention to sequence variants in the regulatory regions that can alter transcription factor binding (Deplancke et al, 2016), affect cell identity (Liu et al, 2017), and bring about disorders like cancer (Killela et al, 2013) and schizophrenia (Fabbri and Serretti, 2017). Genome-Wide Association Studies identify segments that usually contain many sequence variants, of which only one or a few may be directly involved in the development of a disorder. This puts forward the technologies that measure the impact of individual sequence variants directly. Generalization of the limited experimental data for more cell types or functional conditions can be achieved by computational approaches (Shi et al, 2018), e.g., the machine learning methods that can predict the estimated functional impact of the individual variants located in different regulatory elements in the human genome in silico
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have