segmentation fault

[BFeldmeyer]

Hello,
I am running Scaff10x V.4.2 with the following command:
scaff10x -nodes 40 -longread 1 -plot 10x_coverage.png PacBioScaffolds.fasta 19-Bp_S1_L001_R1_001.fastq.gz 19-Bp_S1_L001_R2_001.fastq.gz Result_scaff10x.fasta

Scaff10x is running and also produces a couple of output files, of which align2.dat and try.out are empty:
0 Apr 30 17:40 align2.dat
28G Apr 29 22:56 align.dat
1.3G Apr 29 11:40 cleaN.fasta
2.9M Apr 29 22:56 core.61997
2.5G Apr 29 11:40 tarseq.fastq
19 Apr 29 11:58 tarseq.fastq.amb
1.6M Apr 29 11:58 tarseq.fastq.ann
1.3G Apr 29 11:58 tarseq.fastq.bwt
312M Apr 29 11:58 tarseq.fastq.pac
624M Apr 29 12:06 tarseq.fastq.sa
1.9M Apr 29 11:40 tarseq.tag
0 Apr 29 22:56 try.out

and I obtain the following error message:
...
[M::mem_pestat] low and high boundaries for computing mean and std.dev: (1, 545)
[M::mem_pestat] mean and std.dev: (192.88, 98.24)
[M::mem_pestat] low and high boundaries for proper pairs: (1, 685)
[M::mem_pestat] skip orientation FF
[M::mem_pestat] skip orientation RF
[M::mem_pestat] skip orientation RR
[M::mem_process_seqs] Processed 434416 reads in 662.155 CPU sec, 18.441 real sec
[main] Version: 0.7.17-r1198-dirty
[main] CMD: /cluster/software/scaff10x/Scaff10X-4.2/src/scaff-bin/bwa mem -t 40 tarseq.fastq R1_001.fastq.gz 19-Bp_S1_L001_R2_001.fastq.gz
[main] Real time: 38988.280 sec; CPU: 1509159.415 sec
sh: line 1: 61997 Segmentation fault (core dumped) /cluster/software/scaff10x/Scaff10X-4.2/src/scaff-bin/scaff_bwa -edge 50000 tarseq.tag align.dat align2.dat > try.out
Error running command: /cluster/software/scaff10x/Scaff10X-4.2/src/scaff-bin/scaff_bwa -edge 50000 tarseq.tag align.dat align2.dat > try.out

[zning-sanger]

Thanks for your email. It looks that the code staff_bwa failed. Could you send me a few lines of the texts from align.dat? 10 or two line should be fine. I just want to check if the data format is in line with the pipeline.

Best regards,

Zemin

[zning-sanger]

Sorry 10 or 20 lines from align.dat.

[zning-sanger]

It is also possible that BWA didn't get finished and the number of column in the last line is not equal the one in other lines. If this is the case, you can repeat another run.

[BFeldmeyer]

Hello, here the "head" and "tail" of align.dat *[s_bfeldmeyer@claudius tmp_rununik_14814]$ head align.dat GWNJ-0957:377:GW1902281905:7:1101:17797:1344 83 tarseq_2701 533 23 GWNJ-0957:377:GW1902281905:7:1101:7111:1362 99 tarseq_754 52659 0 GWNJ-0957:377:GW1902281905:7:1101:9546:1362 83 tarseq_24652 18239 0 GWNJ-0957:377:GW1902281905:7:1101:12550:1362 99 tarseq_25146 5604 60 GWNJ-0957:377:GW1902281905:7:1101:14539:1362 83 tarseq_9461 14547 60 GWNJ-0957:377:GW1902281905:7:1101:24383:1362 83 tarseq_7742 42968 14 GWNJ-0957:377:GW1902281905:7:1101:25723:1362 99 tarseq_2070 54217 4 GWNJ-0957:377:GW1902281905:7:1101:8339:1379 99 tarseq_5021 2992 11 GWNJ-0957:377:GW1902281905:7:1101:10064:1379 83 tarseq_1911 54985 17 GWNJ-0957:377:GW1902281905:7:1101:12581:1379 83 tarseq_11715 3540 0 * GWNJ-0957:377:GW1902281905:8:2224:27570:73053 97 tarseq_12548 2460 12 GWNJ-0957:377:GW1902281905:8:2224:28036:73053 83 tarseq_5239 53038 60 GWNJ-0957:377:GW1902281905:8:2224:1336:73071 99 tarseq_12935 14989 60 GWNJ-0957:377:GW1902281905:8:2224:2250:73071 83 tarseq_11559 34264 0 GWNJ-0957:377:GW1902281905:8:2224:2331:73071 99 tarseq_33815 759 40 GWNJ-0957:377:GW1902281905:8:2224:2615:73071 99 tarseq_20916 2188 60 GWNJ-0957:377:GW1902281905:8:2224:2838:73071 99 tarseq_18027 4480 15 GWNJ-0957:377:GW1902281905:8:2224:3143:73071 99 tarseq_20467 16653 60 GWNJ-0957:377:GW1902281905:8:2224:3386:73071 99 tarseq_38403 8267 17 GWNJ-0957:377:GW1902281905:8:2224:3711:73071 99 tarseq_37695 4797 0 Am 04.05.2020 um 16:20 schrieb Zemin Ning:

…

Sorry 10 or 20 lines from align.dat. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOPLJN7AG2APS7RQBHQTI3DRP3FKFANCNFSM4MYV4XAQ>.

[zning-sanger]

Thanks. The file align.dat seems to be fine. How much RAM do you have in your computer? The code of staff_bwa does not use much RAM at this stage, maybe 5-6 GB.

[zhangzhiyangcs]

I have the same question when I ran to scaff_bwa. My RAM is 3TB and my genome is 500Mb. I think it can't be the RAM problem.

[jensbast]

Hi all,
is there a solution in the end?
I am encountering the same problem.
I also have 400 GB memory, so no memory issue.

the align.dat lines:

head align.dat
A00685:102:HWLNMDRXX:1:2101:1217:1000 99 tarseq_73 587650 56
A00685:102:HWLNMDRXX:1:2101:1235:1000 99 tarseq_37 491377 60
A00685:102:HWLNMDRXX:1:2101:1307:1000 83 tarseq_16 526825 60
A00685:102:HWLNMDRXX:1:2101:1976:1000 83 tarseq_44 103907 60
A00685:102:HWLNMDRXX:1:2101:2211:1000 99 tarseq_31 2014651 60
A00685:102:HWLNMDRXX:1:2101:2230:1000 83 tarseq_12 1873831 60
A00685:102:HWLNMDRXX:1:2101:3296:1000 99 tarseq_43 628817 9
A00685:102:HWLNMDRXX:1:2101:3351:1000 83 tarseq_56 766484 60
A00685:102:HWLNMDRXX:1:2101:3477:1000 83 tarseq_76 662403 60
A00685:102:HWLNMDRXX:1:2101:3586:1000 99 tarseq_90 226154 60

tail align.dat
A00685:102:HWLNMDRXX:2:2278:30752:37059 99 tarseq_16 1050079 60
A00685:102:HWLNMDRXX:2:2278:30843:37059 83 tarseq_14 2815439 34
A00685:102:HWLNMDRXX:2:2278:31186:37059 83 tarseq_14 1183768 60
A00685:102:HWLNMDRXX:2:2278:31222:37059 81 tarseq_7 4039605 0
A00685:102:HWLNMDRXX:2:2278:31656:37059 99 tarseq_113 391255 60
A00685:102:HWLNMDRXX:2:2278:31982:37059 83 tarseq_29 451249 60
A00685:102:HWLNMDRXX:2:2278:32072:37059 83 tarseq_93 58721 40
A00685:102:HWLNMDRXX:2:2278:32145:37059 99 tarseq_7 3602339 60
A00685:102:HWLNMDRXX:2:2278:32452:37059 99 tarseq_41 197280 60
A00685:102:HWLNMDRXX:2:2278:32633:37059 65 tarseq_9 2755947 21

[fredjaya]

I encountered the same segfault issue mentioned here and in #18 when using the raw R1 and R2 linked reads as inputs (in this case, TELL-seq reads with barcodes converted to be 10X compatible) for Scaff10X v4.2.

Running scaff_reads to generate debarcoded *.fastq.gz R1 and R2 files, and using these as inputs for scaff10x v4.2 seemed to fix it, as alluded to here.

Note that there may be a path issue with scaff_reads as previously raised (#5) - quick fix here.