Bug: Segmentation fault (core dumped) about dammet

[KHanghoj]

That sounds reasonable. I think/hope that local realignment can only improve the accuracy. I havent explored it. As it is now, the reads only including an Insertion/deletion are just excluded.

Feel free to reopen this issue, if you encounter a similar issue. Or start a new one :)

Best,
kristain

[KHanghoj]

Hi Katterinne,

first of all thanks for the in depth bug report. It seems peculiar, but clearly looks like a bug. Ill investigate.

Thank you providing data to reproduce the bug. it makes it way easier to detect the error. However, it seems like the provided bam is truncated.

samtools quickcheck SF12_DS_05.bam returns SF12_DS_05.bam was missing EOF block when one should be present.

I also tried to jump to chromosome 6: samtools view SF12_DS_05.bam 6 but it didnt work. Would it be possible to send me the bam file again. If the dammet bug can be reproduced baesd on data from chromosome 6, you can just send reads mapping to that.

We will fix this error as soon as we can reproduce it :).

Kristian

When

[Katterinne]

Hi Kristian,

Thanks a lot for the positive and quick reply! I'm so sorry the bam file I provided it's truncated, honestly, don't know what happened there. I'm trying with another sample now, it also produce the error I'm reporting.

This sample also has an average read depth of ~5X, and the chromosomes failing are: 19 and 8.

Download links:
aligment file
bam.bai
read group list file

I hope the file are fine this time, otherwise, please let me know and I'll find another way to share them with you.

Thank you!

Katterinne

[Katterinne]

Hi Kristian,

Just in case you need it and following the alternative you suggested, here is the fastq files with the reads mapping to chromosomes 19 and 8 from the bam file I sent in my last message.

Download links:
reads mapping to chr8
reads mapping to chr19

Thank you!

Katterinne

[KHanghoj]

Hi again,

I have had a look and it seems like some reads in the BAM file is one big insertion. here is an example:

J00121:155:H2FTLBBXY:2:2120:18832:27865 16      19      29379263        47      78I     *       0       0       GGCTCACACACGACGCCCCTGCTCCCTCAGGACCCTGGGCAGAGTCCCGGCTCACACACGACGCCCCTGCTCCCTCAG  FJJJJJJJFJJJFFJJFJJJJJFJJJJFJJJFJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJFFFAA      X0:i:1  X1:i:0  OC:Z:78M        RG:Z:SL22_6_12_L2_R1_001_cMaPZdkPnjey_trim      XG:i:0  NM:i:78 XM:i:3  XO:i:0  XT:A:U

DamMet doesnt know how to handle this so I have added a check so that these reads are removed. chr19 is now running.
But it seems like the BAM file has been modified since mapping. The original cigar is removed (see OC tag with cigar 78M) and replaced by one large insertion.

Let me know if you encounter other problems.

Kristian

[Katterinne]

Hi Kristian,

Sorry for the slow reply; I was excited trying DamMet's upgraded version with the samples that failed before. DamMet estDEAM and estF ran successfully with all the samples, this is great! Thank you very, very much! :D

You're right, i'm sorry I didn't mention it before. We do modify the BAM files after mapping before further analysis (e.g., variant call). What you found, probably happened during the local realignment step. Do you think this could significantly affect DamMet results?

Thanks again! Wish you the best!

Katterinne