You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Background: We want a rare variant analysis module because much of the variance in rare disease incidence is explained by rare mutations. We want to create rare variant methods and use those in publications to push the field of genetics forward.
Background: We want a rare variant analysis module because much of the variance in rare disease incidence is explained by rare mutations. We want to create rare variant methods and use those in publications to push the field of genetics forward.
Task List:
Get underlying statistical model for Dave's simulation written out #249
Generate realistic simulations on Dave's 2 condition simulation #250
Convert Dave's simulation to python and match in terms of summary counts
Write out correct statistical model for genetic correlation with Dave's agreement
To generate data and what we should be modeling
Test the existing models (Reduced Rank Regression, EM ZIP TADA, ML TADA, Dirichlet Process TADA) on Dave's simulations, record performance #251
Add conversion from fitted values to relative risk and heritability #315
Identify best performing model to use as base for multivariate version and communicate results to Dave
Test the same models on our N>3 condition simulation and write down results
Add Bystro Scientific Repo for experimental models that we want to publish on
PR Adversarial model: maximizes explained, minimizes predictive ability on confounding variables. #253
Share results with Dave & ask him about Austin's N condition simulator proposal #254
The text was updated successfully, but these errors were encountered: