Philippos Tsourkas

Assistant Professor


The primary goal of my research is to generate insight into important questions in biology using bioinformatics and computational biology methods. I am also interested in developing bioinformatics and computational tools for use by the biological community. In addition, I am always ready to assist my colleagues with analysis of large biological data sets as the need arises. I am committed to undergraduate research, and all of my publications as an assistant professor at UNLV include undergraduate students as authors. Outlined below are some of the research projects currently pursued in my lab.

Genomics of Paenibacillus larvae bacteriophages

Paenibacillus larvae is a rod-shaped, gram positive bacterium responsible for American Foulbrood Disease (AFB), one of the leading causes of global honeybee population decline. P. larvae form spores that are antibiotic resistant, and currently the only way of containing a P. larvae outbreak is by burning infected hives, causing great financial hardship to beekeepers. Because they pose no risk to honeybees and humans, bacteriophages that infect P. larvae are a potential treatment for AFB. In the last year alone, five studies were published on treating AFB with P. larvae phages or P. larvae phage endolysins, with promising, if not conclusive, results. Furthermore, since 2013, the number of sequence P. larvae bacteriophages has increased from zero to 18, and growing rapidly. In collaboration with Dr. Penny Amy’s lab at UNLV, we have so far sequenced, annotated, and published the genomes of nine P. larvae phages isolated from samples from three US states (with more on the way), and have published the first comparative genomics manuscript on P. larvae phages. As interest in P. larvae phages continues to grow along with the number of sequenced P. larvae phages, key areas we are interested in are: Identifying the function of P. larvae phage proteins, piecing together the genomic diversity and evolutionary history of P. larvae phages, obtaining a full picture of the mechanisms by which P. larvae phages lyse their hosts, and identifying the role of phage-encoded beta-lactamases and toxins in P. larvae antibiotic resistance and virulence. Other potential areas of interest are identifying the mechanism by which P. larvae phages penetrate their host, the relationship of P. larvae phages to their hosts in the wild, including the phages’ role in horizontal gene transfer, identifying uses of P. larvae phage proteins for biotechnology applications, understanding how P. larvae defend against infection from phages, and further studies on the use of P. larvae phages as a treatment for AFB.

A meta-tool for bacteriophage gene prediction and genome annotation

Bacteriophages are the most numerous and diverse entities on Earth, with an estimated 10 31 particles in the biospher. The rapid decrease in cost of sequencing technology has resulted in an explosion in the number of published phage genomes. Consequently, accurate gene prediction and start codon calling (in cases where a gene has more than one possible start codon) for newly assembled phage genomes is of great importance. There are currently several gene calling programs (Glimmer, GeneMark, GeneMark.hmm) that are widely used to annotate new phage genomes in an automated manner. While producing rapid results, such programs occasionally produce false positives and missed calls (false negatives.)  Furthermore, these programs may not necessarily be in agreement in many cases, complicating the process of gene and start codon calling. Accurate genome annotation thus requires manual curation, using additional information such as expert knowledge, BLAST results, gene length, and overlap with other genes (since bacteriophage genes seldom overlap much with each other). However, when the number of sequenced phages is large, manual curation becomes prohibitively time-consuming. To this end we are working towards developing a genome annotation meta-tool for bacteriophages that integrates all readily available information to call genes and start codons, thereby combining as many of the advantages of manual curation as possible, while retaining the speed of automation. We hope to develop our annotation method into freely-available public domain program for use by the phage community.

Identification of gene regulatory networks controlling obesity in starvation selected Drosophila melanogaster

The identification of genes and gene regulatory networks implicated in control of obesity is of great importance in biology and health. In collaboration with the Gibbs lab at UNLV, we are interested in identifying gene regulatory networks that play a central role in regulating obesity in D. melanogaster. Successive generations of D. melanogaster have been subjected to starvation conditions, such that in each generation, only 15% of the individuals survive. In the next generation, the survivors are allowed to breed to replenish the population, and the population is again subjected to starvation conditions. This process has been repeated for >90 generations, resulting in a population of starvation-selected D. melanogaster. D. melanogaster subjected to this kind of starvation selection are prone to obesity when fed a normal diet. We plan to collect gene expression data by RNA-Seq for each generation of starvation selected D. melanogaster and identify genes that are overexpressed in the starvation selected individuals by statistical comparison with a control population. By taking expression profiles at each generation we will have time-course expression data, which we hope to use to infer gene regulatory networks that play a role in obesity.

Investigation into the use of extrinsic kinases in lymphocyte receptor signaling

Signaling by receptors is in many cases mediated by a tyrosine kinase domain that transfers a phosphate group from an ATP molecule to a cytosolic signaling molecule, initiating a cascade that eventually leads to gene transcription. Most commonly, the receptor’s kinase domain is an intrinsic part of the receptor itself (e.g. in the EGFR family of receptors, insulin receptors, etc…). In lymphocytes however (T and B cells), the intracellular domain of the lymphocyte antigen receptor (the receptor dedicated to detecting foreign pathogens, known as the “B cell receptor” or “BCR” in B cells, and the “T cell receptor”, or “TCR” in T cells) does not possess a kinase domain. Rather, kinase activity is carried out by an extrinsic family of molecules known as Src-family kinases that carry out their signaling function by binding to an Immuno-Tyrosine Activation Motif (ITAM) on the intracellular domain of the receptor following antigen ligation to the receptor’s extracellular domain. It is currently not known why lymphocyte antigen receptor signaling differs from other receptor families in this respect. One reason could be that lymphocyte antigen receptors, in contrast to other receptor families, encounter an essentially infinite variety of antigenic ligands. We are currently developing a mathematical model that we hope will generate useful insight into the differences between extrinsic and intrinsic kinase-mediated signaling cascades.


Bioinformatics, Mathematical Modeling


Ph.D. University of California, Berkeley