Prediction and mechanistic dissection of transcriptional activation domains using deep learning and pooled screening
Abstract/Contents
- Abstract
- Transcription factors (TFs) comprise a DNA-binding domain and an effector domain that regulates nearby genes. Activation domains (ADs) -- effector domains that increase transcription -- have long been of particular interest due to their roles as oncogenic drivers and use as scientific tools. We combined quantitative, high-throughput measurements of in vivo activation and in vitro interaction with computational modeling to characterize ADs. Using a domain-tiling in vivo screen, we identified all ADs in budding yeast. We trained a neural network named PADDLE (Predictor of Activation Domains using Deep Learning in Eukaryotes) that accurately predicts ADs across species, and experimentally confirmed predictions of 23 new ADs in human TFs. Surprisingly, and suggesting an expanded role, ADs were also predicted and confirmed in all major coactivator complexes, including Mediator, TFIID, SAGA, cohesin, and condensin. Guided by PADDLE predictions, we designed and measured activation of thousands of AD mutants to derive a deeper understanding of the principles underlying activation. ADs shared no common sequence motifs. Acidic and bulky hydrophobic residues were each necessary for activation, but excess hydrophobicity was strongly inhibiting. While a few ADs required alpha helical folding, most activated based on biochemical features and did not need any specific sequence or secondary structure. We then used mRNA display and pull-down experiments to measure in vitro binding of coactivator proteins to all TF domains. Remarkably, 73% of ADs bound Mediator and binding strength was strongly predictive of activation strength. In contrast, TFIID bound only 18% of ADs, all of which also bound Mediator. Structural modeling revealed that ADs interact with Mediator without shape complementarity ("fuzzy" binding). Kinetic measurements showed that multivalent Mediator-AD interactions were high affinity yet allowed dynamic exchange of bound molecules. We propose that this dynamic exchange is a consequence of fuzzy binding, enables rapid recruitment and release of individual Mediator complexes during activation, and drives phase separation when gene-regulating proteins are at high concentrations.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2021; ©2021 |
| Publication date | 2021; 2021 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Sanborn, Adrian Longson |
|---|---|
| Degree supervisor | Dror, Ron, 1975- |
| Degree supervisor | Kornberg, Roger D |
| Thesis advisor | Dror, Ron, 1975- |
| Thesis advisor | Kornberg, Roger D |
| Thesis advisor | Kundaje, Anshul, 1980- |
| Degree committee member | Kundaje, Anshul, 1980- |
| Associated with | Stanford University, Computer Science Department |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Adrian Longson Sanborn. |
|---|---|
| Note | Submitted to the Computer Science Department. |
| Thesis | Thesis Ph.D. Stanford University 2021. |
| Location | https://purl.stanford.edu/fv392zx5704 |
Access conditions
- Copyright
- © 2021 by Adrian Longson Sanborn
- License
- This work is licensed under a Creative Commons Attribution 3.0 Unported license (CC BY).
Also listed in
Loading usage metrics...