Open frankMusacchia opened 5 years ago
Francesco,
Not sure, but the read depth for that test sample seems incredibly low. I suspect the answer is simply that it is not possible to call CNVs for such a sample?
Dear Vincent, I guess that is the reason. By the way, the error blocked my automated script. I solved creating a blacklist to avoid the analysis of a sample. Best regards Francesco
Il giorno mar 30 apr 2019 alle ore 10:13 Vincent Plagnol < notifications@github.com> ha scritto:
Francesco,
Not sure, but the read depth for that test sample seems incredibly low. I suspect the answer is simply that it is not possible to call CNVs for such a sample?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/vplagnol/ExomeDepth/issues/18#issuecomment-487857157, or mute the thread https://github.com/notifications/unsubscribe-auth/AAOID5RLUFMXXSPYMIR55DTPS75RJANCNFSM4HG5U3IQ .
-- Francesco Musacchia - Ph.D. Genomic Medicine - Telethon Institute of Genetics and Medicine Via Campi Flegrei 34, 80078 Pozzuoli (NA), Italy. Tel. +39 081 19230692 Mobile: +39 349 6396351
Dear Vincent, I am using ExomeDepth for a big batch of BAM file. I am now having this error:
It looks like the test samples has only 2 bins with more than 5 reads. The coverage is too small to perform any meaningful inference so no likelihood will be computed. Error in if ((i > 2) && (res.data.frame$mean.p[i] < 0.05)) break : missing value where TRUE/FALSE needed**
How can I deal with it? The optimization worked for all other files but not for this BAM.
Best Regards, Francesco