Segmentation Fault sorting a BAM file

[kjsalimian]

Hi,

I mapped paired end reads with Bowtie2 to produce a SAM file. Any reason why the following happens???

I run the following commands:

samtools view -bS myfile.sam > myfile.bam
samtools sort myfile.bam myfile.sorted

I get the following output:


[samopen] SAM header is present: 25 sequences.
[bam_sort_core] merging from 87 files...
[bam_header_read] EOF marker is absent. The input is probably truncated.
[bam_header_read] bgzf_check_EOF: Invalid argument
[bam_header_read] invalid BAM binary header (this is not a BAM file).
Segmentation fault

Thanks!

[maubp]

One guess is it is running out of space wherever the temporary files are, resulting in a truncated temporary file and thus the EOF marking missing warning.

[kjsalimian]

It turns out you were right! Thanks for responding. It was due to a memory issue. However, I got a new error this time.
After generating the sam file I ran the following codes, the goal being to create a sam file that is namesorted and containing a region of chr4 only.

samtools view -bS myfile.sam > myfile.bam
samtools sort myfile.bam myfile.sorted
samtools index myfile.sorted.bam
samtools view myfile.sorted.bam 'chr4: 1-1000000' -o myfileextracted.sorted.bam
samtools sort -n myfileextracted.sorted.bam myfile.namesorted
samtools view myfile.namesorted > myfilenamesorted.sam

I got the following errors though:

[bam_sort_core] merging from 87 files...
[bam_header_read] EOF marker is absent. The input is probably truncated.
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[bam_sort_core] truncated file. Continue anyway.
Segmentation fault

It seems to have gone on to try to write the sorted bam file but the size of the sorted bam file is significantly smaller than the original (17.1 gb vs 14.3 gb). Any thoughts on why this may be happening???

[GenoMax]

This has been discussed before. See this thread: http://seqanswers.com/forums/showthread.php?t=13652

It seems to have gone on to try to write the sorted bam file but the size of the sorted bam file is significantly smaller than the original (17.1 gb vs 14.3 gb). Any thoughts on why this may be happening???

[maubp]

A sorted BAM file should be smaller than an unsorted BAM file, however if samtools merge crashed it is quite likely that the file is incomplete.

One simple check would be to count the number of reads in the input and output BAM files.

[Lv Ray]

It turns out you were right! Thanks for responding. It was due to a memory issue. However, I got a new error this time.
After generating the sam file I ran the following codes, the goal being to create a sam file that is namesorted and containing a region of chr4 only.

samtools view -bS myfile.sam > myfile.bam
samtools sort myfile.bam myfile.sorted
samtools index myfile.sorted.bam
samtools view myfile.sorted.bam 'chr4: 1-1000000' -o myfileextracted.sorted.bam
samtools sort -n myfileextracted.sorted.bam myfile.namesorted
samtools view myfile.namesorted > myfilenamesorted.sam

I got the following errors though:

[bam_sort_core] merging from 87 files...
[bam_header_read] EOF marker is absent. The input is probably truncated.
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[bam_sort_core] truncated file. Continue anyway.
Segmentation fault

It seems to have gone on to try to write the sorted bam file but the size of the sorted bam file is significantly smaller than the original (17.1 gb vs 14.3 gb). Any thoughts on why this may be happening???

I got the similar wrong!
I didn't have [bam_header_read] EOF marker is absent. The input is probably truncated.
But I directly got
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[bam_sort_core] truncated file. Continue anyway.
Segmentation fault(core dumped)

could anyone tell me how to deal with this problem! I did another SEQanswers:http://seqanswers.com/forums/showthread.php?t=44875