influence of parameterization on translated transcript alignments

20 views

Skip to first unread message

Olivia

unread,

Apr 25, 2024, 1:53:14 PMApr 25

to rna-star

Hi Alex,

I'm testing how parameters impact STAR alignment and have a question about how genome alignments are translated into transcript alignments when using the --quantMode TranscriptomeSAM option. As an example, the following read has the same genome alignment in Aligned.sortedByCoord.out.bam files generated from three STAR runs with different parameterizations. Unfortunately I did not output the alignment score in test B, this appears to be the only part that differs between A and C.

Genome alignment Aligned.sortedByCoord.out.bam

Parameter test set A (basically ENCODE)

A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 163 SIRV2 4695 255 7S106M988N37M = 4765 1196 CCTACTGGTGTCTGTCGCAATGTAAATGGGGGTGACAAGTTTTACCATTTGGTATGTTTTAGTTTACACATCACACACTATTTCAACTAAACTCGCTACAACGTAGTGAACTCTCGGCATGATATGCTACCTTCTACAATTATTGCTGTT FFFFFFFFFFFFFF:FFFFF,F:FF:FFFFFFFFFFF:FF:FF,FFF:FFFFFFFFFFFFFFFF::FFFF,,:FF,,FFFFFFFFF:FF::FFFFFFFFF:FFFFFFFFF,F:F,FFFFFFF::,,FFF:F,FFFF:FFFFFFFFFFF:F NH:i:1 HI:i:1 AS:i:276 NM:i:2 MD:Z:68A31A42
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 83 SIRV2 4765 255 36M988N102M = 4695 -1196 CTATTTCAACTAAACTCGCTACAACGTAGTGAACTCTCGGCATGATATGCTACCTTCTACAATTATTGCTGTTTCGGTAGGGTTGGATATCATTGCGTATATTTCGAAATGTCGTTCGCATCCATGTTCATCCACCAC :FFFF,FFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFF:F:FFFFFFF:FFFFFF NH:i:1 HI:i:1 AS:i:276 NM:i:1 MD:Z:30A107

Parameter test set B (high multimap, high mismatch)

A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 163 SIRV2 4695 255 7S106M988N37M = 4765 1196 CCTACTGGTGTCTGTCGCAATGTAAATGGGGGTGACAAGTTTTACCATTTGGTATGTTTTAGTTTACACATCACACACTATTTCAACTAAACTCGCTACAACGTAGTGAACTCTCGGCATGATATGCTACCTTCTACAATTATTGCTGTT FFFFFFFFFFFFFF:FFFFF,F:FF:FFFFFFFFFFF:FF:FF,FFF:FFFFFFFFFFFFFFFF::FFFF,,:FF,,FFFFFFFFF:FF::FFFFFFFFF:FFFFFFFFF,F:F,FFFFFFF::,,FFF:F,FFFF:FFFFFFFFFFF:F NH:i:1 HI:i:1 NM:i:2 MD:Z:68A31A42
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 83 SIRV2 4765 255 36M988N102M = 4695 -1196 CTATTTCAACTAAACTCGCTACAACGTAGTGAACTCTCGGCATGATATGCTACCTTCTACAATTATTGCTGTTTCGGTAGGGTTGGATATCATTGCGTATATTTCGAAATGTCGTTCGCATCCATGTTCATCCACCAC :FFFF,FFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFF:F:FFFFFFF:FFFFFF NH:i:1 HI:i:1 NM:i:1 MD:Z:30A107

Parameter test set C (high multimap, med. mismatch, short window)

A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 163 SIRV2 4695 255 7S106M988N37M = 4765 1196 CCTACTGGTGTCTGTCGCAATGTAAATGGGGGTGACAAGTTTTACCATTTGGTATGTTTTAGTTTACACATCACACACTATTTCAACTAAACTCGCTACAACGTAGTGAACTCTCGGCATGATATGCTACCTTCTACAATTATTGCTGTT FFFFFFFFFFFFFF:FFFFF,F:FF:FFFFFFFFFFF:FF:FF,FFF:FFFFFFFFFFFFFFFF::FFFF,,:FF,,FFFFFFFFF:FF::FFFFFFFFF:FFFFFFFFF,F:F,FFFFFFF::,,FFF:F,FFFF:FFFFFFFFFFF:F NH:i:1 HI:i:1 AS:i:274 NM:i:2 MD:Z:68A31A42
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 83 SIRV2 4765 255 36M988N102M = 4695 -1196 CTATTTCAACTAAACTCGCTACAACGTAGTGAACTCTCGGCATGATATGCTACCTTCTACAATTATTGCTGTTTCGGTAGGGTTGGATATCATTGCGTATATTTCGAAATGTCGTTCGCATCCATGTTCATCCACCAC :FFFF,FFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFF:F:FFFFFFF:FFFFFF NH:i:1 HI:i:1 AS:i:274 NM:i:1 MD:Z:30A107

This read appears to derive from the internal standard SIRV gene 2 in a region identical between transcripts SIRV201 and SIRV202. The transcript alignments differ between Aligned.toTranscriptome.out.bam files generated from the three parameter tests.

Transcriptome alignments Aligned.toTranscriptome.out.bam

Parameter test set A

A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 355 SIRV202 22 3 138M = 87 215 GTGGTGGATGAACATGGATGCGAACGACATTTCGAAATATACGCAATGATATCCAACCCTACCGAAACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAG FFFFFF:FFFFFFF:F:FFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFF,FFFF: NH:i:2HI:i:1
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 403 SIRV202 87 3 150M = 22 -215 AACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAGTGTGTGATGTGTAAACTAAAACATACCAAATGGTAAAACTTGTCACCCCCATTTACATTGCGACAGACACCAGTAGG F:FFFFFFFFFFF:FFFF,F:FFF,,::FFFFFFF,F:F,FFFFFFFFF:FFFFFFFFF::FF:FFFFFFFFF,,FF:,,FFFF::FFFFFFFFFFFFFFFF:FFF,FF:FF:FFFFFFFFFFF:FF:F,FFFFF:FFFFFFFFFFFFFF NH:i:2 HI:i:1

Parameter test set B

A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 355 SIRV201 18 3 138M = 83 215 GTGGTGGATGAACATGGATGCGAACGACATTTCGAAATATACGCAATGATATCCAACCCTACCGAAACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAG FFFFFF:FFFFFFF:F:FFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFF,FFFF: NH:i:2HI:i:2
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 403 SIRV201 83 3 150M = 18 -215 AACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAGTGTGTGATGTGTAAACTAAAACATACCAAATGGTAAAACTTGTCACCCCCATTTACATTGCGACAGACACCAGTAGG F:FFFFFFFFFFF:FFFF,F:FFF,,::FFFFFFF,F:F,FFFFFFFFF:FFFFFFFFF::FF:FFFFFFFFF,,FF:,,FFFF::FFFFFFFFFFFFFFFF:FFF,FF:FF:FFFFFFFFFFF:FF:F,FFFFF:FFFFFFFFFFFFFF NH:i:2 HI:i:2

Parameter test set C

A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 99 SIRV202 22 3 138M = 87 215 GTGGTGGATGAACATGGATGCGAACGACATTTCGAAATATACGCAATGATATCCAACCCTACCGAAACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAG FFFFFF:FFFFFFF:F:FFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFF,FFFF: NH:i:2HI:i:1
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 147 SIRV202 87 3 150M = 22 -215 AACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAGTGTGTGATGTGTAAACTAAAACATACCAAATGGTAAAACTTGTCACCCCCATTTACATTGCGACAGACACCAGTAGG F:FFFFFFFFFFF:FFFF,F:FFF,,::FFFFFFF,F:F,FFFFFFFFF:FFFFFFFFF::FF:FFFFFFFFF,,FF:,,FFFF::FFFFFFFFFFFFFFFF:FFF,FF:FF:FFFFFFFFFFF:FF:F,FFFFF:FFFFFFFFFFFFFF NH:i:2 HI:i:1
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 355 SIRV201 18 3 138M = 83 215 GTGGTGGATGAACATGGATGCGAACGACATTTCGAAATATACGCAATGATATCCAACCCTACCGAAACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAG FFFFFF:FFFFFFF:F:FFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFF,FFFF: NH:i:2HI:i:2
A01488:219:HCC2LDSX7:1:2216:5602:2268:AGACGTTTAT 403 SIRV201 83 3 150M = 18 -215 AACAGCAATAATTGTAGAAGGTAGCATATCATGCCGAGAGTTCACTACGTTGTAGCGAGTTTAGTTGAAATAGTGTGTGATGTGTAAACTAAAACATACCAAATGGTAAAACTTGTCACCCCCATTTACATTGCGACAGACACCAGTAGG F:FFFFFFFFFFF:FFFF,F:FFF,,::FFFFFFF,F:F,FFFFFFFFF:FFFFFFFFF::FF:FFFFFFFFF,,FF:,,FFFF::FFFFFFFFFFFFFFFF:FFF,FF:FF:FFFFFFFFFFF:FF:F,FFFFF:FFFFFFFFFFFFFF NH:i:2 HI:i:2

What information is being used to determine the transcript alignment of this read? Is it dependent on the parameterization of the STAR call? I came across this question while trying to better understand how RSEM interprets STAR alignment and quality information to generate counts. Thank you for any advice you can offer.