Predicting promising cancer vaccine targets, pancreatic cancer types drive drug responses, a broken gene behind male infertility, and more.
Research Roundup: December 20, 2019
Welcome to the December 20, 2019 installment of Research Roundup, a recurring snapshot of recent studies published by scientists at the Broad Institute and their collaborators.
Improved prediction of tumor-specific antigens
Predicting tumor-specific antigens from a cancer patient—to inform cancer vaccine development and other therapeutic efforts—is challenging due to diversity between patients of both tumor proteins and the immune system's HLA alleles. Siranush Sarkizova, Susan Klaeger, Karl Clauser, Derin Keskin, institute scientist and Proteomics Platform director Steven Carr, institute member Catherine Wu in the Cancer Program, institute member and Cell Circuits Program co-director Nir Hacohen, and colleagues used a machine learning approach to analyze 186,464 HLA-associated peptides from cell lines representing 95 common and rare HLA types worldwide. The resulting model predicted at least 50 percent more endogenously presented antigens as previous methods, and identified more than 75 percent of the antigens displayed on 11 patient tumor cell lines. Read more in Nature Biotechnology and a Broad news story.
DNA editor for continuous mutagenesis
Haiqi Chen, Schmidt fellow Fei Chen, and their colleagues have developed a DNA editor capable of rapid and efficient mutagenesis in human cells. Their system, called TRACE, consists of a cytidine deaminase fused to a T7 RNA polymerase, and continuously induces nucleotide diversification downstream of an inserted T7 promoter. The team showed that TRACE could rapidly edit within a large window at least 2,000 base pairs long, and used their system to identify functionally correlated mutations in a MEK1 inhibitor-resistance screen. The researchers say that TRACE can be expanded to use other base-editing enzymes and polymerases. Find out more in Nature Biotechnology.
Pancreatic cancer cells' drug responses depend on where they start
Evidence is starting to suggest that pancreatic ductal adenocarcinoma (PDAC) tumor cells might be able to shift between two clinically and transcriptionally distinct identities, dubbed epithelial (E) and quasi-mesechymal (QM). Seeking to understand the therapeutic implications, David Ting (MGH), associate member Daniel Haber, and colleagues examined how the transcriptional states of PDAC cells from lab models and patient tumors change in response to different treatments. They found that the FOLFIRINOX combination regimen generally pushes PDAC cells to be more QM-like, while cells' responses to vitamin D differ greatly depending on their initial subtype. Learn more in PNAS.
Genetic defects may underlie the one in five cases of infertility that are “unexplained.” A team led by institute member Cynthia Morton in the Program in Medical and Population Genetics, Amy MacQueen (Wesleyan), Frank Tuttelmann (U. Munster), and Samantha Schilit (Harvard) examined cases of unexplained male fertility to look for chromosomal rearrangements that disrupt genes important in fertility. In one case, a balanced reciprocal translocation led to dysregulation of the SYCP2 gene near one of the rearrangement breakpoints, and three other cases harbored frameshift variants in SYCP2. Appearing in the American Journal of Human Genetics, the work suggests new causes for infertility in men harboring a balanced chromosomal rearrangement.