Fathoming Cancer by Interrogating Integrated Interactomes
summary
Integrates diverse omics, interaction assays and tissue annotation into a network resource with 18,000 proteins and ~800,000 edges and flags prediction confidence per edge based on evidence types and tissues. Curated-checking of ~200 pairs reveals roughly 50% annotation errors in some databases. Shortest‑path centrality picks 20 lung adenocarcinoma candidates of which 18 of 20 validate prognostically. An optimization picks minimal drug combinations that outperform single drugs in 3 lung cancer cell lines and partially rewire the disease subnetwork.
AI-written summary · about
chapters (15)
- 0:00Intro
- 0:43Integrated Computational Biology
- 2:32Considerations
- 7:08Followup Example
- 9:02Good or Bad
- 9:51Identification of Mechanism
- 10:26Relevant Biology
- 12:51New Interaction Databases
- 13:59Validation
- 16:00Integration
- 25:24Optimal Settings
- 28:45Everyone Lies
- 29:00Ground Truth
- 30:58Diversity is Necessary
- 31:55Heterogeneity
related talks
31mDiverse High Throughput Technologies in Cancer Research and Synthetic Biology
16mUsing Network Analysis to Detect Kickback Schemes Among Medical Providers
45mComplexity in Living Systems
01h 51mMorning Session: Genomics, Medicine, Epidemiology and Data Interpretation
05mTranslating clinical guidance to actionable insights with R
27mTech Talk
27mMining social media data of people with rare diseases
06mHealthcare Analytics in the Age of Personalized Medicine