Projects/MASSIV: Alternative splicing-inspired protein development

From Research management course
Revision as of 18:03, 24 September 2020 by Wiki (talk | contribs)
Jump to: navigation, search

Problem 74

  • Title: MASSIV: Alternative splicing-inspired protein desdign
  • Problem: Alternative splicing and alternative initiation/termination transcription sites, have the potential to greatly expand the proteome in eukaryotes by producing several transcript isoforms from the same gene. Although these mechanisms are well described at the genomic level, little is known about their contribution to protein evolution and their impact at the protein structure level. Here, we address both issues by reconstructing the evolutionary history of transcripts and by modeling the tertiary structures of the corresponding protein isoforms.
  • Data: Almost ready to share, could be accessed on request.
  • References:
    1. Transcripts’ evolutionary history and structural dynamics give mechanistic insights into the functional diversity of the JNK family by Elodie Laine et al, 2020, biorxiv, github
    2. Disentangle homology relationships between exons by LCQB UPMC, 2020, github
    3. Protein physics: a course of lectures with color and stereoscopic illustrations and problems by Finkelstein A.V., Ptitsyn O.B. book in rus
    4. Image www.lcqb.upmc.fr/laine/images/phylo.png
  • Basic solution: for now the problem is not strictly formulated. But there is a github project to start with.
  • Novelty: Eukaryotes have evolveda transcription machinery that can augment the protein repertoire without increasing the genome size. It produces several mRNA transcripts from the same gene, by choosing different initiation/termination sites and/or by splicing different exons. This regulatory process is called Alternative Splicing (AS). Although the mechanisms of AS have been well described at the genomic level, the extent to which and how AS modulates protein functions and interactions remains a fundamental open question. Here, we propose to develop a probabilistic model that will learn from the functional AS events observed today in nature to generate new protein functional diversity. The rationale will be that the means AS has produced to generate protein diversity along evolution can be reused and generalized to expand the protein repertoire way beyond what we observe today. The challenge here will be to design an artificial system able to learn the underlying rules of (functional) AS and to generalize to any protein sequence.
  • Authors: Elodie Laine, Sergei Grudinin, and Vadim Strijov