search

FertigLab / CoGAPS

Bayesian MCMC matrix factorization algorithm
https://www.bioconductor.org/packages/release/bioc/html/CoGAPS.html
BSD 3-Clause "New" or "Revised" License
61 stars 17 forks source link

result interpretation #66

Closed dm8000 closed 3 months ago

dm8000 commented 1 year ago

Hello

I ran the vignette but I have some troubles on interpreting the results.

  1. The rows in the example matrix have a prefix Hs. followed by a number. Is this a gene annotation?

  2. If I understood correctly, would the patterns be groups of correlated features? If so, where can I find which feature belongs to which pattern? I was not able to find it in the result object.

dimalvovs commented 3 months ago

It's been some time, but still:

  1. Hs symbols correspond to the Unigene names, which still can be mapped to other names, for example Hs.93485
  2. Features in patterns are not necessarily correlated, features x patterns matrix can be obtained by calling res@featureLoadings on the resulting CoGAPS object.