Predictive article recommendation using natural language processing and machine learning to support evidence updates in domain-specific knowledge graphs.

Objectives: Describe an augmented intelligence approach to facilitate the update of evidence for associations in knowledge graphs.

Methods: New publications are filtered through multiple machine learning study classifiers, and filtered publications are combined with articles already included as evidence in the knowledge graph. The corpus is then subjected to named entity recognition, semantic dictionary mapping, term vector space modeling, pairwise similarity, and focal entity match to identify highly related publications.

Identification of targetable BRAF ΔN486_P490 variant by whole-genome sequencing leading to dabrafenib-induced remission of a -mutant pancreatic adenocarcinoma.

The tumor genome of a patient with advanced pancreatic cancer was sequenced to identify potential therapeutic targetable mutations after standard of care failed to produce any significant overall response. Matched tumor-normal whole-genome sequencing revealed somatic mutations in , , , and a focal deletion of The variant was an in-frame deletion mutation (ΔN486_P490), which had been previously demonstrated to be a kinase-activating alteration in the BRAF kinase domain. Working with the Novartis patient assistance program allowed us to treat the patient with the BRAF inhibitor, dabrafenib.

Evaluating Clinical Genome Sequence Analysis by Watson for Genomics.

Oncologists increasingly rely on clinical genome sequencing to pursue effective, molecularly targeted therapies. This study assesses the validity and utility of the artificial intelligence Watson for Genomics (WfG) for analyzing clinical sequencing results. This study identified patients with solid tumors who participated in in-house genome sequencing projects at a single cancer specialty hospital between April 2013 and October 2016.