Using Machine Learning and Natural Language Processing to Review and Classify the Medical Literature on Cancer Susceptibility Genes.

Purpose: The medical literature relevant to germline genetics is growing exponentially. Clinicians need tools that help to monitor and prioritize the literature to understand the clinical implications of pathogenic genetic variants. We developed and evaluated two machine learning models to classify abstracts as relevant to the penetrance-risk of cancer for germline mutation carriers-or prevalence of germline genetic mutations.

Artificial Anti-Tumor Opsonizing Proteins with Fibronectin Scaffolds Engineered for Specificity to Each of the Murine FcγR Types.

We have engineered a panel of novel Fn3 scaffold-based proteins that bind with high specificity and affinity to each of the individual mouse Fcγ receptors (mFcγR). These binders were expressed as fusions to anti-tumor antigen single-chain antibodies and mouse serum albumin, creating opsonizing agents that invoke only a single mFcγR response rather than the broader activity of natural Fc isotypes, as well as all previously reported Fc mutants. This panel isolated the capability of each of the four mFcγRs to contribute to macrophage phagocytosis of opsonized tumor cells and in vivo tumor growth control with these monospecific opsonizing fusion proteins.