STAR GenomeIndex output (which is the BAM file?)

Question

[daniloceschin]

Hi all,

I'm using STAR to generate a Genome Index to then use it for a Genome-guided Trinity analysis. The parameter I have used are:

STAR --runThreadN 20 \
     --runMode genomeGenerate \
     --genomeDir tibGenIdex/ \
     --genomeFastaFiles nanPar1.fa \
     --genomeChrBinNbits 10 \
     --outSAMtype BAM SortedByCoordinate \
     --sjdbGTFfile tibetan.gtf \

The output files are:

115K  - chrLength.txt
457K - chrNameLength.txt
343K - chrName.txt
271K - chrStart.txt
9.1M - exonGeTrInfo.tab
3.8M - exonInfo.tab
635K - geneInfo.tab
2.0G - Genome
567 - genomeParameters.txt
16G - SA
1.5G - SAindex
6.4M - sjdbInfo.txt
5.9M - sjdbList.fromGTF.out.tab
5.9M - sjdbList.out.tab
2.6M - transcriptInfo.tab

Which is supposed to be the BAM file?

There are huge files: Genome: 2.0G; SA: 16G and SAindex: 1.5G. Should I maintain those files?

Thank you all for any help!

Best regards,

[GenoMax]

This produces a genome index which is used for alignment of regular fastq data independent of Trinity. You can't use this index for Trinity directly.

From Trinity support page.

Users must provide read alignments to Trinity as a coordinate-sorted bam file. Use GSNAP, TopHat, STAR or other favorite RNA-Seq read alignment tool to generate the bam file, and be sure it's coordinate sorted by running 'samtools sort' on it.

[daniloceschin]

Thank you! Now I understand how it works. You have made me think! :)

Thank you all for your fast replay.

Answer 1

[yhoogstrate]

genomeGenerate doesn't prdoce a SAM/BAM file, STAR only produces one if you run it in normal mode upon some FASTQ/FASTA data and the generated index.

[Ram]

Note: "normal" as in "alignReads", which is the default.