search

im3sanger / dndscv

dN/dS methods to quantify selection in cancer and somatic evolution
GNU General Public License v3.0
212 stars 48 forks source link

columns qmis_cv, qtrunc_cv, qallsubs_cv, pglobal_cv, qglobal_cv NA #49

Closed guidmt closed 4 years ago

guidmt commented 4 years ago

Dear Dr. Marticorena,

I am trying to use your useful package, however, I get results in which the columns qmis_cv, qtrunc_cv, qallsubs_cv, pglobal_cv, qglobal_cv are "NAs". The only column with some values is pind_cv.

Screenshot 2020-05-27 at 22 38 31

I get also these warnings:

1: In dndscv(parse_maf_file, refdb = "/mnt/data/lab/gmt_data/pseudospace2/dNdScv/dndscv_data-master/data/RefCDS_human_GRCh38.p12.rda", : Same mutations observed in different sampleIDs. Please verify that these are independent events and remove duplicates otherwise. 2: In model$y == genemuts$n_syn : longer object length is not a multiple of shorter object length

edit: 28/05/2020 I have done some tests, using the parameters cv=NULL I have resolved the "problem".

Many thanks in advance for your support,

Guidantonio

im3sanger commented 4 years ago

Thank you, Guidantonio. Yes, I was going to suggest that. There are no covariates for GRCh38 yet (apologies), so "cv=NULL" needs to be used. Users can also generate their own covariates (e.g. coverage per gene in their data, expression level, etc).

Best wishes, Inigo