Training Program

2017-19 Cancer Systems Biology Trainees

Beril Tutuncuoglu, a postdoctoral fellow in Nevan Krogan’s lab (UCSF). This project sought to find additional synthetic lethal targets for olaparib by performing a CRISPR/Cas9-based chemogenetic screen. Along with identifying known factors in homologous recombination, this screen identified four new factors whose absence increased olaparib sensitivity. The data obtained here formed the basis of a successful F32 fellowship.

Fan Zheng, a postdoctoral fellow in Trey Ideker’s lab (UC San Diego). This project used both CCMI-generated and publicly available data to build a hierarchical model of cellular subsystems in human cells. This hierarchy was then used to identify 300 mutational “bottlenecks,” subsystems under significant positive selection even if the individual genes are not, across 13 cancer cohorts. Some are now undergoing experimental validation.

Jackie Einstein, a graduate student in Gene Yeo’s lab (UC San Diego). This project aimed to identify RNA binding protein (RPB) networks that control cancer cell survival using focused genomic screens to identify synthetic lethal RBPs. A Myc synthetic lethality screen revealed YTHDF2, an m6A reader protein, as a mRNA destabilizer. Several members of the same pathway were candidate hits in an independent KRAS synthetic lethality screen.

Juan Jado Rodriguez, a postdoctoral fellow in Elizabeth Winzeler’s lab (UC San Diego). This project sought to identify mechanisms underlying resistance to cancer therapies by generating drug-resistant human haploid cell lines followed by whole exome sequencing. The drug sensitivities of candidate genes are currently being tested using CRISPR/Cas9-based approach in both haploid and other human cell lines.

Kivil Ozturk, a graduate student in Hannah Carter’s lab (UC San Diego). This project aimed to develop bioinformatics tools to improve driver mutation classification by modelling the impact of cancer mutations on PPI network architectures. The network-features based classifier outperformed both classifiers that use only features from the state-of-the-art missense mutation prioritization tool VEST, or the combined features.

Mehdi Bouhaddou, a postdoctoral fellow in Nevan Krogan’s lab (UCSF). The computational project integrated proteomic data from cell lines and patients using network propagation to identify proteins associated with cetuximab-resistance in head and neck cancer. This created a resource map of altered protein-protein interactions and revealed a protein subnetwork signature of drug resistance. This data helped support a successful F32 application.

Michael Liao, a graduate student in Jeff Hasty’s lab (UC San Diego). This project aimed to improve therapeutic stability of synthetically engineered bacteria that preferentially grow in disease-affected environments, using a three-strain ecology where each strain can kill and replace the previous strain. Cyclical addition of strains to a common cell culture allowed for a prolonged therapeutic window in the absence of antibiotics. A mathematical model describing switching population dynamics further permitted analytical predictions.

Michelle Moritz, a specialist in David Agard’s lab (UCSF). This project aimed to gain an atomic-level understanding of how regulatory casein kinase-1δ binds to the γ-Tubulin Small Complex (γTuSC) using cross linking mass spectrometry and cryo-EM. After testing many conditions, Michelle found that Spc110-γTuSC complexes assemble into stable helical filaments that should allow for high resolution structural determination.