Statistical Bioinformatics Seminar: Signal -- Machine learning annotation of branchpoints and in silico modelling of functional splicing events

The aim of the statistical bioinformatics seminar is to provide a forum for 
people working within the broad area of computation and statistics and their 
application to various aspects of biology to present their work and showcase 
their ongoing projects. It is intended to foster the exchange of ideas and 
build potential collaborations across multiple disciplines.


The seminars will be held at 1:00 pm on Monday in Charles Perkins Centre 
Seminar Room (Level 3, large meeting room). The format of the talk is 30~45 
minutes plus questions.


Speaker: Beth Signal (Garvan Institute of Medical Research)

Title: Machine learning annotation of branchpoints and in silico modelling 
of functional splicing events.

Abstract: 
RNA splicing is a key component of mature RNA transcript formation, required for the 
removal of intronic regions and subsequent ligation of exonic regions. This process 
can also allow for alternative splicing to occur, where different exonic regions are 
ligated together to produce alternative RNA products. 

The branchpoint element is one of the splicing sequence elements, required for the 
first lariat-forming reaction in splicing. However current catalogues of human 
branchpoints remain incomplete due to the difficulty in experimentally identifying 
these elements. To address this limitation, we have developed a machine-learning 
algorithm - branchpointer - to identify branchpoint elements solely from gene 
annotations and genomic sequence. Using branchpointer, we annotate branchpoint 
elements in 85% of human gene introns with sensitivity (61.8%) and specificity 
(97.8%). In addition to annotation, branchpointer can evaluate the impact of 
SNPs on branchpoint architecture to inform functional interpretation of genetic 
variants. Branchpointer identifies all published deleterious branchpoint mutations 
annotated in clinical variant databases, and finds thousands of additional clinical 
and common genetic variants with similar predicted effects. 

While alternative splicing can produce alternative RNA products, a large 
proportion of these have little functional impact on open reading frames or 
transcript stability. To address this limitation in the functional interpretation 
of differential splicing analyses, we have developed software to model events in 
silico and interpret their functional impact. 


About the speaker: Beth is a PhD Student in the Clinical Genome Informatics group 
at the Garvan Institute. Her current research is focused on developing 
bioinformatics methods to understand how transcript splicing and expression is 
controlled. She has a particular interest in using machine learning techniques 
to study transcriptomic behaviour.