Problems with STAR mapping output

Question

[manekineko]

Hi,

I have problems seeing the output from STAR mapper. I have done the mapping, then made an sorted bam for IGV but I see very few reads mapped, and I have this log:

 Number of input reads |       116876253
                      Average input read length |       20
                                    UNIQUE READS:
                   Uniquely mapped reads number |       93912191
                        Uniquely mapped reads % |       80.35%
                          Average mapped length |       19.85
                       Number of splices: Total |       129488
            Number of splices: Annotated (sjdb) |       0
                       Number of splices: GT/AG |       129455
                       Number of splices: GC/AG |       33
                       Number of splices: AT/AC |       0
               Number of splices: Non-canonical |       0
                      Mismatch rate per base, % |       0.00%
                         Deletion rate per base |       0.00%
                        Deletion average length |       1.00
                        Insertion rate per base |       0.00%
                       Insertion average length |       1.00
                             MULTI-MAPPING READS:
        Number of reads mapped to multiple loci |       20734865
             % of reads mapped to multiple loci |       17.74%
        Number of reads mapped to too many loci |       2072050
             % of reads mapped to too many loci |       1.77%
                                  UNMAPPED READS:
       % of reads unmapped: too many mismatches |       0.00%
                 % of reads unmapped: too short |       0.00%
                     % of reads unmapped: other |       0.13%

I'm particularly interested to see the reads that are splitted in exon-intron sites:

Number of splices: GT/AG |       129455

I have noticed that when converting from SAM to BAM the size decreased from 16GB to only 60MB which is strange?!

[Ashutosh Pandey]

Well 16 GB to 60 Mb is strange. Have you checked if your BAM file is correct? Try running samtools view Input.bam and see if it throws EOF marker is absent error.

[mark.ziemann]

In addition to Ashutosh's recommendation, can you do a

samtools view -c input.bam

to count the rows in the bam and

samtools view -c -q10 input.bam

to count the uniquely mapped reads. If you have fewer than 93912191 reads then the sam->bam conversion failed somehow.

[manekineko]

it returned error:

samtools view -c -q10 Aligned.out.bam
[bam_header_read] EOF marker is absent. The input is probably truncated.
[main_samview] truncated file.

but I made the conversion as I always did (of course I tried STAR for the first time, but it should output a standart SAM?)

I did it with -

samtools view -b -S SAM > BAM

[Ashutosh Pandey]

Yes, almost all the aligner these days will output standard SAM. I think there was a problem while converting your sam file to bam file. You should check the log. It could be insufficient space on your system, or a malformed line in the sam file. The error EOF marker is absent suggests that the bam conversion process terminated before the complete conversion.

Answer 1

[mark.ziemann]

STAR 2.4 can output alignments directly in binary BAM format, so I'd recommend an update.

https://github.com/alexdobin/STAR/blob/master/doc/STARmanual.pdf