Gean Hu Rose Hills
Studying activation domain activity using protein structure prediction
Transcription factors (TFs) are essential proteins which control gene expression. Misregulation of TFs is the main driver of many cancers, neurological disorders, and cardiovascular diseases, making the study of TF interactions extremely important. While the DNA-binding regions of TFs have been extensively characterized, the coactivator-binding regions of TFs, termed activation domains (ADs), remain relatively understudied. Previous work here in the Staller Lab showed that orthologs (evolutionarily related proteins from different species) of transcription factor Gcn4 are able to activate gene expression in yeast. This intriguingly happens despite the orthologs sharing minimal similarity in their amino acid sequences, but it is unclear whether these sequences have similar or varied interaction mechanisms with coactivator proteins. In this project, we aim to computationally predict the structure of and analyze Gcn4-coactivator protein complexes using deep learning models. We hope this analysis will uncover new structural rules governing activation domain activity and provide valuable insight for therapeutic strategies targeting transcription factors.
Message To Sponsor
Thank you so much for your generous support of this project! This opportunity to work on meaningful and impactful research in computational biology is extremely valuable to me and my journey as a scientist. I'm very excited to learn about and contribute to the intersection of machine learning and structural biology!