small output bug in starsolo 2.7.10a

[chubukov]

Looks to me like probably the wrong field is being output in Solo.out/Gene/summary.csv

Same input.
STAR-2.7.9a:

Reads Mapped to Gene: Unique+Multipe Gene,0.484805
Reads Mapped to Gene: Unique Gene,0.453594

STAR-2.7.10a_alpha_220314

Reads Mapped to Gene: Unique+Multiple Gene,0.453594
Reads Mapped to Gene: Unique Gene,0.453594

[alexdobin]

Hi Victor,

thanks for catching it, looks like a bug introduced in 2.7.10a.
What parameters did you run with?

Thanks,
Alex

[chubukov]

Hi Alex,

Here's a typical run. One thing that might be relevant is that I built the genome with 2.7.9. Should it be necessary to rebuild it?

/home/vchubukov/STAR-2.7.10a_alpha_220314/source/STAR   --genomeDir /net/seq/data2/projects/prime_seq/genome   --readFilesIn /net/seq/data2/seque
ncers/211222_A01347_0078_AHMTNTDRXY/fastq/CGTACTAG_S2_L002_R2_001.fastq.gz /net/seq/data2/sequencers/211222_A01347_0078_AHMTNTDRXY/fastq/CGTACTAG
_S2_L002_R1_001.fastq.gz   --soloType CB_UMI_Simple   --soloCellReadStats Standard   --clip3pAdapterSeq AAAAAAAAAA   --clip3pAdapterMMp 0.1   --s
oloCBstart 1   --soloCBlen 12   --soloUMIstart 13   --soloUMIlen 16   --soloCBwhitelist /net/seq/data2/projects/prime_seq/barcodes-combined.txt  
 --soloCellFilter  EmptyDrops_CR 72 .99 10 45000 90000 100000 0.01 20000 0.01 10000   --quantMode "TranscriptomeSAM"   --outSAMunmapped Within   
--outFilterType BySJout   --outSAMattributes NH HI AS NM MD CR CY UR UY GX GN CB UB   --outFilterMultimapNmax 20     --outFilterMismatchNmax 999 
    --outFilterMismatchNoverReadLmax 0.04     --alignIntronMin 20     --alignIntronMax 1000000     --alignMatesGapMax 1000000     --alignSJoverha
ngMin 8     --alignSJDBoverhangMin 1   --sjdbScore 1   --readFilesCommand zcat   --runThreadN 10   --limitBAMsortRAM 30000000000   --outSAMtype B
AM SortedByCoordinate   --outSAMheaderCommentFile commentslong.txt   --outSAMheaderHD '@HD' 'VN:1.4' 'SO:coordinate'

[alexdobin]

Hi Victor,

after checking the code I realized this is correct behavior, as multi-gene reads are only counted if you specify --soloMultiMappers options. Without these options there is no difference between Unique and Unique+Multiple Gene counts.
To avoid confusion, in the next release I will replace the Unique+Multiple value with the "NoMulti" string.

Thanks again for pointing it out.
Alex