Predicting peptide antigens, and profiling the infant microbiome.
Research Roundup: October 26, 2018
Welcome to the October 26, 2018 installment of Research Roundup, a recurring snapshot of recent studies published by scientists at the Broad Institute and their collaborators.
Peptide presentation prognosticator produced
We do not fully understand the immune system's rules for determining whether a peptide from a bacterium and other source will 1) be presented to T cells, and 2) spark an immune response. To help bring new clarity, a team led by Daniel Graham, Chengwei Luo, and core institute member Ramnik Xavier from the Broad’s Infectious Disease and Microbiome Program (IDMP) profiled the "peptidome" of potential antigens bound to MHC class II (a protein complex that presents peptides to T cells) in mice. They used their data to build and train BOTA, a machine learning algorithm that predicts antigenic peptides based on bacterial whole genome data. Learn more in Nature Medicine and in a Broad news story.
Watching baby's microbiome grow up
In search of microbial triggers for type 1 diabetes (T1D), a team led by the IDMP's Xavier, Tommi Vatanen, and Curtis Huttenhower analyzed nearly 11,000 metagenomes in stool samples from children at risk for T1D, collected monthly starting at three months of age. Known as The Environmental Determinants of Diabetes in the Young (TEDDY) study, it produced the most shotgun metagenomic microbiome profiles published for a single target population to date. Appearing in Nature, the work finds that the microbiome gains adult-like functions as early as one year of age, and suggests that short-chain fatty acids may protect against early-onset T1D.