Abstract
Transcription factors (TFs) can bind DNA in a cooperative manner, enabling a mutual increase in occupancy. Through this type of interaction, alternative binding sites can be preferentially bound in different tissues to regulate tissue-specific expression programmes. Recently, deep learning models have become state-of-the-art in various pattern analysis tasks, including applications in the field of genomics. We therefore investigate the application of convolutional neural network (CNN) models to the discovery of sequence features determining cooperative and differential TF binding across tissues. We analyse ChIP-seq data from MEIS, TFs which are broadly expressed across mouse branchial arches, and HOXA2, which is expressed in the second and more posterior branchial arches. By developing models predictive of MEIS differential binding in all three tissues, we are able to accurately predict HOXA2 co-binding sites. We evaluate transfer-like and multitask approaches to regularizing the high-dimensional classification task with a larger regression dataset, allowing for the creation of deeper and more accurate models. We test the performance of perturbation and gradient-based attribution methods in identifying the HOXA2 sites from differential MEIS data. Our results show that deep regularized models significantly outperform shallow CNNs as well as k-mer methods in the discovery of tissue-specific sites bound in vivo.
Highlights
Chromatin immunoprecipitation followed by sequencing (ChIP-seq) can reveal the genomic regions bound by transcription factor (TF) proteins in different tissues or developmental stages
In this work we introduced convolutional neural network (CNN) methods for identification of DNA sequence features predicting differential and cooperative TF binding
Validation with HOXA2 ChIP-seq showed that CNN models trained on MEIS data could reliably identify HOXA2 features in BA2, consistent with a synergistic effect of HOXA2 and MEIS binding [8]
Summary
Chromatin immunoprecipitation followed by sequencing (ChIP-seq) can reveal the genomic regions bound by transcription factor (TF) proteins in different tissues or developmental stages. Short DNA reads are aligned to a reference genome assembly and peak calling techniques such as MACS [1] are used to localise the regions enriched in the IP experiment compared to a control. Inferred TF peak locations are typically hundreds to thousands of base-pairs in length and contain functional sequence motifs identifiable as highly over-represented short k-mers or position-specific score matrices (sequence motifs, usually 6-10nt), corresponding to the binding locations of regulatory TFs. Widely used motif discovery tools include MEME [2], Homer [3], GEM [4] and KSM [5].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
