Transcription factor binding site detection using convolutional neural networks with a functional group-based data representation

Abstract

Transcription factors (TFs) play an essential role in molecular biology by regulating gene expression. The binding sites of TFs can vary by a large amount and the numerous possible binding locations make their detection a challenging issue. Recently, several machine learning approaches using nucleotide sequence data were applied to classify DNA sequences regarding Transcription Factor Binding Sites (TFBS). We propose a novel training strategy without the traditional 1D nucleotide-based DNA sequence representation by instead using a 2D topological matrix of sub-nucleotide chemical functional groups substantially defining the protein binding ability of DNA fragments. We train convolutional neural networks using this novel Functional Group DNA Representation (FGDR) to solve a TFBS classification task. We compare our results with the efficiency of previous nucleotide-based training approaches and show that learning from an FGDR data sequence has several benefits regarding TFBS classification. Moreover, we reason that learning deep neural networks from the FGDR representation produces competitive results while only introducing a pre-processing conversion step. Finally, we show that employing an ensemble of models from the nucleotide and FGDR representations for network training results in higher classification performance than any of the single input approaches.

Modeling methyl-sensitive transcription factor motifs with an expanded epigenetic alphabet

“…For representing expanded epigenetic alphabet data, we would suggest using an approach where significantly lower frequency of modified bases compared to unmodified has a lower impact. One might instead encode all nucleobases as vectors representing their functional groups, as recently done in a similar unmodified context [ 95 ]. Alternatively, one might adapt recent work which designed filters to create sparse codes, similar to images, that can effectively encode DNA motifs [ 96 ].…”

Computational approaches to understand transcription regulation in development

“…While large compendia of TF binding profiles in different cell types have been collected for humans [37], this effort remains unfeasible for less well-studied organisms, including most developmental model systems. With sufficient training data, TF binding can be computationally imputed [50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65], however, this does not necessarily generalize across species [66]. As a result, most current approaches use relatively simple models that combine experimentally measured CRE activity with TF binding motifs to computationally predict TF binding.…”