Predicting novel transcription factor-target gene interactions in the Candida albicans biofilm network using machine learning

Transcription is a complex process underlying many cellular functions. DNA structure was discovered by Rosalind Franklin over 50 years ago. Since then, we have been steadily dissecting our understanding of the biological logic governing all life on Earth. Ultimately, Francis Crick discovered the central dogma of molecular biology in 1970. Early studies on Escherichia coli began formulating the idea of gene regulation as the basis of information flow in biological systems. Relatively soon, due to advancements in computing power, computer aided analyses of DNA and RNA were introduced to understand regulatory sequences. Since then, many improvements in experimental and computation protocols have accelerated our understanding of the control of biological information flow.My thesis work, which uses the fungal species Candida albicans as a model, is the latest advancement in our understanding of gene regulation. In chapter one, I present my work on identifying the gene regulatory networks controlling biofilm formation in C. albicans across the biofilm life cycle. I explain my novel workflow that utilizes sequence-based as well as DNA-shape based features to predict transcription factor binding sites (TFBSs) genome-wide in C. albicans. In chapter two, I present CUT&RUN sequencing data implemented to assess binding events for specific TFs in C. albicans. I also present my novel CUT&RUN computational pipeline to analyze CUT&RUN sequencing data. In chapter three, I present a computational workflow to analyze 3’ Tag-Seq data. In this 3’ Tag-Seq, the 3’ end of the transcript is selected and amplified to yield one copy of cDNA from each transcript in a biological sample