How Developers can treat Ovarian Cancer
summary
An evidence engine that turns patient sequencing into treatment recommendations is already built and deployed: it parses ~0.5 TB per genome, normalizes public and clinic data into a Neo4j graph of over a million nodes, and produces ranked implications linked to line-of-evidence so clinicians can verify suggestions. The pipeline uses Clojure for implication logic and macros, Neo4J for the graph store and JRuby for parsing, goes to production on AWS with Puppet, and ships in about three months, moving Clearity from ~10 patients/month toward scale.
AI-written summary · about
chapters (15)
- 0:00Start
- 0:35Overview
- 1:02Key Players
- 2:54Concepts of Genomics and Cancer
- 3:10What Is a Genome
- 7:39Why Are We Learning about Dna
- 9:54Facts about Ovarian Cancer
- 10:38Standard of Care
- 20:19Neo4j
- 24:41Summary
- 25:42Learnings
- 27:58Feedback and Learning
- 29:13Legal Implications
- 32:49How Big Was the Treatment Option Data Set
- 35:29Cancer Genome Atlas Data
related talks
35mUsing Large-Scale Clinico-Genomics Data for in silico Clinical Trials and Precision Oncology
50mCancer Genomics - a biologist and a developer
20mTranslating a Trillion Pointo Diseases of Data into Therapies
12mHow data science can make your genome actionable
25mUsing Python Biopharm to Help Cure Cancer
11mEnsuring Reliability in Healthcare Data Engineering
10mGenesis Final Round - Drylab AI
50mThe algorithm for precision medicine