The availability of full genome sequences for many organisms has greatly increased the reach of bioinformatics. In my research, I have used a variety of techniques to leverage the information carried in mouse, human, and viral genomes to address a diverse set of challenges.
One challenge was to devise a set of sequences to detect various strains of Human Papillomavirus (HPV). Chapter I describes the method by which I designed probe sequences common to multiple genomes to efficiently isolate HPV DNA from human tissue samples and probe sequences unique to each HPV genome to differentiate between viral strains for the purpose of diagnosing infections.
Chapter II depicts my role in developing the prototype International Gene Trap Consortium web resource, which presents information about embryonic stem cell lines carrying single gene knockouts to the public. Much of this work involved the creation of a new web site and a multi-path process for identification of gene trap sequence tags. Chapter III describes work that developed out of the transition from an mRNA transcript-based sequence tag annotation method to a process that combines transcript matching with localization to the mouse genome. To understand better the localization of gene trap sequence tags to the mouse genome, I compared stand-alone versions of the common genome alignment programs BLAT, SSAHA, and MegaBLAST.
Chapter IV details a method to detect splice variation in different tissues. I developed a process to combine information about splice variants gained by aligning expressed-sequence tags (ESTs) with full-length gene transcripts with microarray analysis to detect splice variants in high-throughput expression data. This method utilized data from pre-existing microarray expression experiments, and so had the potential for large-scale academic and industry use.