search

pneuvial / c3co

Inferring cancer cell clonality from copy-number data
5 stars 1 forks source link

Centering data and normal tumor subtype modelling #65

Open jchiquet opened 5 years ago

jchiquet commented 5 years ago

At this point, we do not explicitly use a normal subtype in the model. This might bet handle latter with something like an intercept term in the model:

The data centering before the call to initializeZt() is a leftover of an attempt made in the intercept branch, which now make now sense. Indeed as, shown by the in the consistency test for initializeZt() in commit b2e4d23c37688db8c43b27c58218aca7c4eda56f, centering the data does not affect the estimation as long as forceNormal = FALSE (which is the default).

Now, forceNormal = TRUE (adding a column of 1) does not make sense anymore since positiveFusedLasso() does not account for a normal sub-clone in its current form. We should then remove the centering and forceNormal in initializeZt().

pneuvial commented 5 years ago

Decision: this should not be addressed at initialization but above.

This is/will be addressed by #68