We are made up of hundreds of different cell types carrying out a diverse range of functions essential for organism survival. All the information required to specify the morphology, function and response to stimuli of these cells is encoded in identical copies of the genome. The process of gene regulation determines where and when different combinations of genes are transcribed.
Given the pivotal role of transcription factors (TFs) in gene regulation it is surprising how little we know about TFs. Here we aim to combine existing transcriptome datasets covering the majority of human tissues and primary cells to delineate the function of transcription factors.
Our novel approach relies on competitively training hierarchal dirichlet process hidden Markov models (HDPHMM) on a complete set of cell type specific data. Preliminary data suggests that this approach has the potential to reveal important biological signals undiscoverable when looking at individual samples.
This research is funded by a BHP Blue Sky grant.