Statistical Bioinformatics Seminar Series: Duren

Title: Continuous lifelong learning for modelling of gene regulation from single cell
multiome data by leveraging atlas-scale external data 

Speaker: Dr Zhana Duren (Clemson University) 

Abstract: Inferring context-specific Gene Regulatory Networks (GRNs) from genomics data
is a crucial task in computational biology.  However, the accuracy of inferred GRNs is
often low due to the limitations of current methods.  We developed a method called
scPECA, which infers gene regulation from single cell Paired gene Expression and
Chromatin Accessibility data from the same cell.  We also propose a metric called the
"pioneer index" which aims to improve the accuracy of GRN and interpretability of the
model by providing a quantitative measure of the TFs’ ability to initiate chromatin
remodeling.  The scPECA method achieved 3 times higher accuracy compared to currently
available GRN inference methods when ChIP-seq data was used as the ground truth.  We
found that disease genes from both differential expression analysis and Genome-Wide
Association Studies (GWAS) were enriched in the target genes of TFs with high pioneer
index scores.  

About the speaker: Dr Zhana Duren earned his BS in Mathematics and Applied Mathematics
from Beihang University (China) in 2012.  He was awarded his PhD in Operational Research
and Cybernetics from Academy of mathematics and systems science, Chinese Academy of
Sciences.  From 2015-2020, he worked in Professor Wing Hung Wong’s lab at Stanford
University as a visiting PhD student (2015-2017) and postdoc research fellow
(2017-2020).  He is currently Assistant Professor in the Department of Genetics and
Biochemistry at Clemson University.