DeepDiff: DEEP-learning for predicting DIFFerential gene expression from histone modifications

Abstract

Motivation Computational methods that predict differential gene expression from histone modification signals are highly desirable for understanding how histone modifications control the functional heterogeneity of cells through influencing differential gene regulation. Recent studies either failed to capture combinatorial effects on differential prediction or primarily only focused on cell type-specific analysis. In this paper we develop a novel attention-based deep learning architecture, DeepDiff, that provides a unified and end-to-end solution to model and to interpret how dependencies among histone modifications control the differential patterns of gene regulation. DeepDiff uses a hierarchy of multiple Long Short-Term Memory (LSTM) modules to encode the spatial structure of input signals and to model how various histone modifications cooperate automatically. We introduce and train two levels of attention jointly with the target prediction, enabling DeepDiff to attend differentially to relevant modifications and to locate important genome positions for each modification. Additionally, DeepDiff introduces a novel deep-learning based multi-task formulation to use the cell-type-specific gene expression predictions as auxiliary tasks, encouraging richer feature embeddings in our primary task of differential expression prediction. Results Using data from Roadmap Epigenomics Project (REMC) for ten different pairs of cell types, we show that DeepDiff significantly outperforms the state-of-the-art baselines for differential gene expression prediction. The learned attention weights are validated by observations from previous studies about how epigenetic mechanisms connect to differential gene expression. Availability and implementation Codes and results are available at deepchrome.org. Supplementary information Supplementary data are available at Bioinformatics online.

Recursive Convolutional Neural Networks for Epigenomics

“…This strategy can also be applied to different organisms with varying number of chromosomes and varying chromosome lengths. For each gene in the dataset, we focused on a genome context of maximum 15,000 bases upstream and downstream of the Transcription Start Site (TSS), thus remaining within the range of reported short range interactions between histone modifications and gene expression ( [38], [39]). Within our implementation the histone signal is calculated in bins of size 1 base or higher, without restrictions on the size of the bins.…”

Deep Learning in Omics Data Analysis and Precision Medicine

“…Using CNNs, Yin et al designed an algorithm to predict these histone modifications by integrating sequence and DNase data (44). In addition, Singh et al used a CNN to infer gene expression from histone modifications data (45), while Sekhon et al used a LSTM to predict differential gene expression, also from histone modifications data (46).…”

Iterative Epigenomic Analyses in the Same Single Cell

“…In bulk cell analysis, characterization of multiple histone modifications predicted gene expression more effectively than characterization of single histone modifications (Karlic et al 2010; Dong and Weng 2013;Singh et al 2016;Sekhon et al 2018;Yin et al 2019). In addition to capturing concurrent patterns of gene expression, broad epigenomic profiling in bulk cells could predict patterns of gene expression and cell phenotype in response to environmental stimuli (Bock et al 2011;Krausgruber et al 2020).…”