Some Computer Studies of Membrane Proteins, Molecular Chaperones, and Color

Citation

Saladi, Shyam Madhukar (2024) Some Computer Studies of Membrane Proteins, Molecular Chaperones, and Color. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/40cw-kn70. https://resolver.caltech.edu/CaltechTHESIS:10232023-184021847

Abstract

This thesis shares a series of stories on seemingly disparate topics united by my efforts and love of computers. Initially, I discuss how the challenge of membrane protein expression provided an initial impetus for research. I channeled efforts towards developing a predictive (machine-learning) model for heterologous overexpression in E. coli. While we made strides to extend this model to other systems (not discussed here), my time was refocused onto questions of more fundamental biochemical interest: the biogenesis of tail-anchored membrane proteins. I built structural, predictive, and phylogenetic models to better understand how the C-terminal domain of co-chaperone Sgt2 functioned, refined the definition of the wider Sti1 family which includes Sgt2-C, and extended our understanding of those features of tail-anchored proteins that determine successful targeting in Yeast and Human cells. I developed a deep phylogeny of Get3, a chaperone involved in tail-anchored protein biogenesis, and helped specifically place Get3 proteins of photosynthesising organisms into evolutionary context. Along the way, I developed a parallel and compelling theme around data visualization, specifically around the use of colormaps across the life sciences. In particular, I built an application to screen and notify preprint authors when their manuscript had poor colormap usage. This was the first time automated software has been used to help authors improve their work at the preprint stage, an area that has grown significantly since my initial work. Finally, I brought together structural biology and data visualization by making perceptually uniform colormaps available in popular molecular visualization software tools to advocate for more thoughtful color usage in the field.

Thesis Files