Genome-wide chromatin accessibility prediction with deep neural networks
Abstract/Contents
- Abstract
- Transcriptional regulation is closely associated with chromatin accessibility. Critical regulatory elements such as enhancers and repressors undergo a very dynamic chromatin structural variation to initiate gene regulation during cell differentiation. Large collaborative consortiums have widely applied the assays of measuring genome-wide chromatin accessibility onto hundreds of human cell types to investigate the mechanism of cell differentiation. Many integrative analyses focus on identifying epigenetic changes of chromatin structure that can lead to differential gene expression. However, it remains a mystery how chromatin accessibility is governed in the first place. To address this problem, we present a standard pipeline of processing sequencing data from the two most commonly adopted chromatin accessibility assays, DNase-seq and ATAC-seq, and then introduce an approach of deep neural networks (DNNs) to learn the underlying rules of chromatin regulation and predict open genomic regions in a wide variety of human cell types, using only the gene expression of transcriptional factors (TFs). We train DNNs on 682,425 enhancer related regulatory regions with 620 TFs that have sequence motifs in 100 cell conditions, and demonstrate high accuracy of prediction. Our model shows great generalizability on new cell conditions that it has not been trained on. The learned coefficients of our input features reveal useful insights into the functional roles of TFs. Therefore, our DNN offers an interpretable model of transcriptional regulation of chromatin accessibility.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2017 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Fu, Xing Margaret |
|---|---|
| Associated with | Stanford University, Institute for Computational and Mathematical Engineering. |
| Primary advisor | Wong, Wing Hung |
| Thesis advisor | Wong, Wing Hung |
| Thesis advisor | Hastie, Trevor |
| Thesis advisor | Saunders, Michael |
| Advisor | Hastie, Trevor |
| Advisor | Saunders, Michael |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Xing Margaret Fu. |
|---|---|
| Note | Submitted to the Institute for Computational and Mathematical Engineering. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2017. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2017 by Xing Fu
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...