LARGE SCALE COMPUTATIONAL ANALYSIS OF THE IMPACT OF ALTERNATIVE SPLICING ON THE REGULATORY FUNCTION OF TRANSCRIPTION FACTORS AND THEIR PROTEIN DOMAINS

Abstract

Transcription factors (TFs) are proteins that bind to specific genomic regions to drive, regulate, or repress gene expression. Alternative splicing is a process that allows a single gene to generate multiple mRNA transcripts, each of which can potentially be translated into distinct protein isoforms with different functions. This process can lead to the production of various TF splice isoforms with different structural and functional properties. The regulation of TFs through alternative splicing can impact gene expression and cellular processes in several ways: 1) Differential Functions (i.e. TF isoforms that have variations in their protein domains, which may lead to different DNA-binding specificities, affinities, or interaction partners); 2) Tissue- or Stage-Specific Expression (i.e. certain TF isoforms may be active in specific developmental stages or in particular tissues, allowing for fine-tuned gene regulation in different contexts); 3) Disease and Pathway Regulation: (i.e. aberrant splicing of TFs can contribute to disease states, including cancer).

This project provides an overview of the main alternative splicing scenarios in human TF isoforms, their TF family superclass, the domains present and the possible impact they may have on general TF activity. By using publicly available databases, I analysed 8427 isoforms from 1406 TF genes with multiple protein coding isoform. The impact of AS on protein domain architecture, including DNA-binding domains (DBDs) and 14 novel categories of non-DBD ‘partner’ domains was systematically analysed using bioinformatic tools and databases such as DIGGER and ISOGO. 27% of TF isoforms exhibited no domain number changes, while 22% lacked at least one DBD, potentially impacting DNA binding affinity. 8% lacked at least one partner domain, potentially affecting protein-protein interactions and gene regulatory functions. 1% of isoforms gained domains. Additionally, 4% exhibited domain truncations, and 2% exhibited domain extensions, potentially altering protein structure and function.

Analysis of GTEx RNA-seq data revealed substantial heterogeneity in TF isoform expression across 30 human tissues. Expression patterns varied widely, with some tissues exhibiting dominant isoforms (over 80% proportional expression), others exhibiting alternative isoform dominance considered as a switched event, and others displaying more even expression across isoforms – distributed. The analysis did not reveal any consistent associations between protein domains or tissue types and the observed patterns of dominant, switched, or distributed isoform expression.

This work lays a foundation for functional studies offering profound insights into the dynamic roles played by TFs and more importantly their isoforms. The results from this study can be applied more broadly, for example in cancer studies, to highlight the importance of isoform specificity as we develop treatments in personalised medicine.

Date of Award	1 May 2025
Original language	English
Awarding Institution	King's College London
Supervisor	Alessandra Vigilante (Supervisor) & Jeremy Green (Supervisor)