Skip to content

Releases: BoevaLab/ONCOCNV

OncoCNV v7.0

02 Nov 08:59
826c24e
Compare
Choose a tag to compare

OncoCNV v7.0

OncoCNV v6.9 - more bugs fixed

28 Sep 19:33
Compare
Choose a tag to compare
Fix rare clustering error

OncoCNV v6.8

24 Mar 17:29
Compare
Choose a tag to compare

Better support for "chr" in fasta files and BAM

OncoCNV v6.7

23 Mar 16:14
Compare
Choose a tag to compare

Should not throw an error during the t.test when there are NA data; Ln. 1240 of processSamples.R

ONCOCNV v6.6

05 Aug 15:56
Compare
Choose a tag to compare

processSamples.v2.1.R - version where gene-based p-values are taken into account
processSamples.v2.2.R - version where gene-based p-values are taken into account only if a segmentation breakpoint falls within the gene
processSamples.v2.3.R - version with T-test to get higher p-values
processSamples.v5.0.R,processControl.v5.0.R - version with GC-correction, amplicon-based counts and amplicon length correction
processControl.v5.2.R - uses RCA and Loess instead of ICA and cubic splines
processControl.v5.3.R - sghseg added as an option for profile segmentation, some bugs fixed
processControl.v5.4.R - possiblity to choose number of PCs in PCA is added. Less than 3 PCs for less than 4 controls
processControl.v5.5.R - can treat samples without chrX/Y and "bad" amplicons with zero read counts (bug fixed)
processSamples.v5.6.R - serious bug fixed: losses and gains can be inverted in the previous versions; normalization of profiles with big amplification is improved
processControl.v5.7.R - can process control files with just one sample, in this case it says that PC1=normalized read counts for the control sample
ONCOCNV_getCounts.v5.9.pl - bug with GOTO is solved
processSamples.v6.0.R - "t.test for two identical values" bug solved
processSamples.v6.1.R - does not create an error when no information is available for no amplicons of a given gene
processControl.v6.1.R - does not create NA for the smallest amplicon length if there are more than 5% of such amplicons (change this parameters in the file if needed)
processControl.v6.2.R - bug in v6.1 is corrected - for the smallest aplicon length
ONCOCNV_getCounts.v6.2.pl - does not look for overlaps of regions when there are more than 50000 targeted regions. Otherwise, it takes ages if one does not call bedtools or use Set::IntSpan..
ONCOCNV_getCounts.v6.3.pl - can read .BAM files where there is no "chr" prefix in from of each chromosome name
processSamples.v6.4.R - fixed a bug when there are a few aplicons with an extremely low variance in control; and no variance at all in tumors
processSamples.v6.5.R - can now find losses with 0 reads in several amplicons in a row; can output copy number for chrY and chrM.
processSamples.v6.6.R - does not throw an error when there are large regions with zero reads in the control (this can happen only when there are serious experemental problems).
processControl.v6.6.R - can now process exome-seq data with a file with prob locations (preferably. Otherwise use exons). - Will still take several hours.