SBME Research Seminar - Dr. Hannah Carter
Immune Checkpoint Blockade (ICB) has revolutionized cancer treatment, however mechanisms determining patient response remain poorly understood. We used machine learning to predict ICB response from germline and somatic biomarkers and studied feature usage by the learned model to uncover putative mechanisms driving superior outcomes. Patients with higher T follicular helper infiltrates were robust to defects in the class-I Major Histocompatibility Complex (MHC-I). Further investigation uncovered different ICB responses in MHC-I versus MHC-II neoantigen reliant tumors across patients. Despite similar response rates, MHC-II reliant responses were associated with significantly longer durable clinical benefit (Discovery: Median OS=63.6 vs. 34.5 months P=0.0074; Validation: Median OS=37.5 vs. 33.1 months, P=0.040). Characteristics of the tumor immune microenvironment reflected MHC neoantigen reliance, and analysis of immune checkpoints revealed LAG3 as a potential target in MHC-II but not MHC-I reliant responses. This study highlights the value of interpretable machine learning models in elucidating the biological basis of therapy responses.
Events
Calendar
- This event has passed.
SBME Seminar: Towards a Context-Aware Atlas of Human Variant Effects – Dr. Frederick Roth
October 16, 2024 @ 12:00 pm - 1:00 pm PDT
SBME Seminar: Towards a Context-Aware Atlas of Human Variant Effects – Dr. Frederick Roth
Seminar Abstract:
Although computational and experimental advances have enabled large-scale determination of sequence variant effects, the real-world impacts of pathogenic variants often depend on environmental or genetic context. For multiple human proteins, e.g., MTHFR and LDLR, I will highlight the potential for large-scale cell-based assays, carried out under different contexts, to model the human phenotypic impact of missense variants. I will also review computational variant effect predictors, describe a new way to assess performance that avoids ascertainment and circularity biases, and convey the need for predictors to become context aware. Together, I hope these topics will convey that a new field of context-aware variant effect mapping is emerging at the interface of experimental and computational biology.
Dr. Frederick Roth’s Biography
Roth trained in physics and biology at UC Berkeley, in biophysics at Harvard, and worked with two biotech companies. He has led research teams at Harvard Medical School, the University of Toronto, and now Chairs the Department of Computational and Systems Biology at the University of Pittsburgh. His team developed now-common computational methods to identify functional enrichment and transcription factor binding sites from transcriptome data, and carried out many analyses of protein and genetic networks. His group is now focused on large-scale mapping of dynamic protein interactions and on the context-dependent impacts of human sequence variation.
Location:
DMCBH 101 LT