Open malcook opened 3 years ago
Hi Malcolm,
yes, the behavior was changed in 2.7.8a, and it is mentioned in the CHANGES.md:
--clip*
now require specifying the values for all read mates, even if they are identical.
I know such changes break people's pipeline and I try to minimize them. But sine I was reworking the clipping code, I realize the previous behavior was somewhat confusing.Cheers Alex
Ok, thanks, but,
readFilesIn=’a1.1.fq,a2.1.fq,a3.1.fq a1.2.fq,a2.2.fq,a3.2.fq’
should I then have
clip3pAdapterSeq="CTGTCTCTTATACACATCT,CTGTCTCTTATACACATCT,CTGTCTCTTATACACATCT”
Or maybe space delimited?
clip3pAdapterSeq="CTGTCTCTTATACACATCT CTGTCTCTTATACACATCT CTGTCTCTTATACACATCT”
Or maybe I need to specify it for both reads in the read pair, as
clip3pAdapterSeq="CTGTCTCTTATACACATCT,CTGTCTCTTATACACATCT,CTGTCTCTTATACACATCT CTGTCTCTTATACACATCT,CTGTCTCTTATACACATCT,CTGTCTCTTATACACATCT”
??
I’m guessing it’s the first (being the one I didn’t try yet).
Hi Malcolm,
will fix the manual, thanks! The values are space-separated and specified for read-1 and read-2 only. If you have multiple files for each mate, these values will be considered the same for all files.
Cheers alex
Hi Alex,
I have the exact same problem and I'm not entirely sure how to specify the adapter seq with --clip3pAdapterSeq . I tried comma-separated, space-separated, without and quotes and all of them return the same error
STAR --runThreadN 20 --limitBAMsortRAM 10000000000 --runMode alignReads --genomeDir ~/share/mm10_STAR --outSAMtype BAM SortedByCoordinate --clip3pAdapterSeq "AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC" --outReadsUnmapped Fastx --quantMode GeneCounts --outFilterType BySJout --outSAMattributes NH HI AS NM MD --outFilterMismatchNmax 999 --outFilterMismatchNoverReadLmax 0.04 --outFilterScoreMinOverLread 0.4 --outFilterMatchNminOverLread 0.4 --alignIntronMin 20 --alignIntronMax 1000000 --alignSJoverhangMin 8 --alignSJDBoverhangMin 1 --sjdbScore 1 --alignMatesGapMax 1000000 --readFilesCommand zcat --readFilesIn fastq/5-1264_R1_001.fastq.gz fastq/5-1264_R2_001.fastq.gz --outFileNamePrefix Klf5/STAR/
EXITING because of fatal PARAMETER error: --clip3pAdapterSeq has to contain 2 values to match the number of mates.
SOLUTION: specify 2values in --clip3pAdapterSeq , for no clipping use -
Thanks Apoorva
specifying --clip3pAdapterMMp parameter along with --clip3pAdapterSeq fixed the issue
Hi Apporva,
also, I would remove quotes:
--clip3pAdapterSeq AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC
,
otherwise, it may consider it as one sequence.
Cheers Alex
I'm rerunning a pipeline after upgrading STAR to 2.9.4 and getting this error which used not to arise for the identical job.
It is unclear why this might now happen.
There is nothing in recent release notes that seems to pertain.
And the manual continues to read:
re-running but with 2 values for clip3pAdapterSeq,
--clip3pAdapterSeq CTGTCTCTTATACACATCT,CTGTCTCTTATACACATCT
continues to exit with same message.Full output log from second run follows: