Uniquely mapped read, but not primary alignment?

[Alex Ng]

Hi Alex and group,

I'm using STAR 2.4.0j to align reads to the human genome (GRCh38), using --clip5pNbases 9 (for UMIs) and --quantMode TranscriptomeSAM (to get transcripts named in the SAM/BAM).

The problem I have is no reads are considered primary alignments (i.e. all reads, regardless of whether it is multi-mapping or not, has the "not primary alignment" flag. See below.) The default should automatically set the single best alignment as primary right? Am I doing something wrong? 

Thanks, Alex

M00455:558:000000000-ALRWY:1:1102:18757:5762    419     ENST00000508493 23      255     9S67M   =       187     231     AATTCCGGTGCGAGATGACGAAGGGAACG

TCATCGTTTGGAAAGCGTCGCAATAAGACGCACACGTTGTGCCGCCG    ACCCAFFGGGGGGGGFFGFGGCGEGGGGGGGGCGGGGGGGGGGFG7@CFGGGGGGGGFGGGGGGGFGGGGFGGEGG    NH:i:1  HI:i:1

M00455:558:000000000-ALRWY:1:1102:18757:5762    339     ENST00000508493 187     255     67M9S   =       23      -231    AAAAGACGAAATACCACCGGAACTGGTCG

AATGAGGCACCTAAAAATTGTATACCGCAGATTCAGCGATCGACAAC    GGFGFDGGGGGGGGGFDGGGFGGGGGGCGFGGGGGGGGFFEGGGGGGGGGFGFDGGGGGFDEEGGDGGGGGCCCCC    NH:i:1  HI:i:1

M00455:558:000000000-ALRWY:1:1102:6342:5763     419     ENST00000376772 174     255     9S67M   =       343     236     AGGACATGTCGGAGGCCCAGGAGGAGGGG

GAATAAAGAAGGTGGAGGATCCTGGCTACCACTCTGAATCCGATACC    CCCCCGGGGGGGGGGGGGGGGFGGGGGGGGGGGGGGGGGGGGGGGGGGFGGGGGGGGGGGGGGGGGGGGGGGGGGG    NH:i:1  HI:i:1

M00455:558:000000000-ALRWY:1:1102:6342:5763     339     ENST00000376772 343     255     67M9S   =       174     -236    GGCTGAGCGAGACCGTCACTGAGACAACG

GTTACCGTGACAACCGAGCCCGAGAACCGGAGCCTTACATCCAAGCA    GGGGGGGGGDGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGCCCCC    NH:i:1  HI:i:1

[Alexander Dobin]

Hi Alex,

I believe this problem was fixed in the later release, please try the latest one.

Cheers

Alex