paperKB
coga / coga-kb
Help
Sign in

Chunk #10 — Results — Methods overview. — Step 1a:

Source
Flexible statistical methods for estimating and testing effects in genomic studies with multiple conditions.
Embedded
yes

Text

Generate candidate covariance matrices U = (U1,…, UK). This list includes both “data-driven” matrices that are estimated from the strongest signals in the condition-by-condition results (see Online Methods), and “canonical” matrices that have simple interpretations. For example, the “canonical” matrices include the identity matrix (representing independent effects across all conditions); a matrix of all ones (representing effects that are equal in all conditions); and R matrices that represent effects that are specific to each condition.