shilpagarg / pstools Goto Github PK

./pstools haplotype_scaffold -t32 scaff_connections.txt pred_haplotypes.fa ./
start main
All above 5M: 0
All above 1.5M: 0
38020 Segmentation fault ./pstools haplotype_scaffold -t32 scaff_connections.txt pred_haplotypes.fa ./

Hi I tried this as suggested shilpagarg/DipAsm#16 in 1.5 TB of RAM and with 72, 64 and 32 cores
I also tried running hifiasm with other parameters like -l 3 and -s 0.45 but I always get the same segmentation fault running in CentOS Linux release 7.4.1708 (Core)

I see 3 other people have asked this same question before, is there a fix for this?
#5
#6
#4
Thanks

Just for your information.
Here is the log: https://github.com/junaruga/pstools/runs/4163849768?check_suite_focus=true

Sorry to interrupt you.
I have run the commands as you recommended in the readme.md. Here are my commands which got the Segmentation fault (core dumped)

pstools hic_mapping -t100 -o pstools.map C010-CHA-E10_hifiasm.dip.r_utg.fa hic_r1.fq.gz hic_r2.fq.gz
pstools resolve_haplotypes -t100 -i true pstools.map C010-CHA-E10_hifiasm.dip.r_utg.gfa pstools

where C010-CHA-E10_hifiasm.dip.r_utg.gfa is 8g resulted from hifiasm(trio-binning), and pstools.map is only 100m(I don't know whether it's approrate or not).

Best.

Hi,

It is a very interesting concept you present in pstools. I would be very curious to see phasing performance for particular genome. Unfortunately, resolve_haplotypes consistently fails at a same point:

...
Start get Sequence: utg019112l to utg023847l
Finish get Sequence: utg014729l to utg016802l
Finish get Sequence: utg014646l to utg016393l
Finish get Sequence: utg016785l to utg023311l
Finish get Sequence: utg007839l to utg014248l
Finish get Sequence: utg006195l to utg004256l
Finish get Sequence: utg009564l to utg014552l
Finish get Sequence: utg008379l to utg012250l
Finish get Sequence: utg007720l to utg008464l
Finish get Sequence: utg013148l to utg023083l
Finish get Sequence: utg008626l to utg019008l
Finish get Sequence: utg019112l to utg023847l
Finish get Sequence: utg005218l to utg009287l
Finish get Sequence: utg002365l to utg006658l
Finish get Sequence: utg007585l to utg008971l
Finish get Sequence: utg012274l to utg015527l
Finish get Sequence: utg015688l to utg017541l
Finish get Sequence: utg017089l to utg021840l
Finish get Sequence: utg003477l to utg005574l
Finish get Sequence: utg010467l to utg021432l
Finish get Sequence: utg012512l to utg016870l
Finish get Sequence: utg005659l to utg016905l
Finish get Sequence: utg015729l to utg027229l
Finish get Sequence: utg010244l to utg013714l
All above 10M: 0
All above 1.5M: 6
Segmentation fault (core dumped)

Since there is very little/none information reported it is difficult to trace the reason.

I can give more background on data, assembly, hardware and etc, but perhaps that "Segmentation fault" error rings the bell immediately?

Cheers,
Iggy

Hi,

I was interested in running pstools with Omni-C data but I had the next error segementation fault error running hic_mapping_haplo.

... [M::tb_pipeline::10452.864*42.22] processed 1333334 sequences; [M::tb_pipeline::10462.076*42.23] processed 1333334 sequences; [M::tb_pipeline::10470.184*42.24] processed 1333334 sequences; [M::tb_pipeline::10474.139*42.25] processed 671892 sequences; /var/spool/slurmd/job106890/slurm_script: line 11: 12235 Segmentation fault (core dumped) ./pstools_1 hic_mapping_haplo -t64 02.4_Nypro_Flye_sspace_lr1_gc1_SLRr1_gc1.scaff_seqs <(zcat /cluster/home/shared/nypro/08.8_ALLHiC/fastq/RacoonDogHiC_EKDL200002762-1a_R1.fastq.gz) <(zcat /cluster/home/shared/nypro/08.8_ALLHiC/fastq/RacoonDogHiC_EKDL200002762-1a_R2.fastq.gz) -o scaff_connections.txt /var/spool/slurmd/job106890/slurm_script: line 12: 48458 Killed

"hic_mapping_haplo" ran out of RAM with a 1 TB server. I tried to follow shilpagarg/DipAsm#16 like you suggested in a previous issue but it didn't work.

Do you know how could I solve it?

Cheers,
Luis

Hi, I followed shilpagarg/DipAsm#16 like you suggested for human genome assembly.
./pstools_1 resolve_haplotypes -t32 hic_name_connection.output NA24385.asm.r_utg.gfa ./ works well on NA24385.

But when I use the same command on NA12878: ./pstools_1 resolve_haplotypes -t32 hic_name_connection.output NA12878.asm.r_utg.gfa ./
it leads to this error:

start main
start get bubbles
Segmentation fault (core dumped)

Environment:

• OS: CentOS 8.2 (64-bit)
• CPU

Could you please help me solve this problem? Thank you very much.

i got a error during make process, the error was as follow: hic_mapping.cpp: 320:11 error: 'idx' does not name a type for (auto idx: counts) . i test it on several linux platform with different g++ version

dear author: i've installed g++ and zlib, but an error was reported when I installed pstools:

/home/miniconda3/bin/x86_64-conda-linux-gnu-c++ -fvisibility-inlines-hidden -std=c++17 -fmessage-length=0 -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -ffunction-sections -pipe -isystem /home/miniconda3/include -DNDEBUG -D_FORTIFY_SOURCE=2 -O2 -isystem /home/miniconda3/include -c -o hic_switch_error.o hic_switch_error.cpp
hic_switch_error.cpp: In function 'void get_switch_error(pldat_t*, bseq_file_t*, bseq_file_t*, std::map<std::__cxx11::basic_string, std::__cxx11::basic_string >)':
hic_switch_error.cpp:538:31: error: unable to find string literal operator 'operator""PRIu64' with 'const char [13]', 'long unsigned int' arguments
printf("Switch at %"PRIu64"\n", condensed_counting[p].first);
^~~~
hic_switch_error.cpp:541:33: error: unable to find string literal operator 'operator""PRIu64' with 'const char [15]', 'long unsigned int' arguments
printf("Unswitch at %"PRIu64"\n", condensed_counting[p].first);
^~~~
hic_switch_error.cpp:564:31: error: unable to find string literal operator 'operator""PRIu64' with 'const char [15]', 'long unsigned int' arguments
printf("Unswitched: %"PRIu64"\n", total_vars - hamming_distance);
^~~~
hic_switch_error.cpp:565:29: error: unable to find string literal operator 'operator""PRIu64' with 'const char [13]', 'long unsigned int' arguments
printf("Switched: %"PRIu64"\n", hamming_distance);
^~~~
hic_switch_error.cpp:567:40: error: unable to find string literal operator 'operator""PRIu64' with 'const char [38]', 'long unsigned int' arguments
printf("Switch error number: %"PRIu64", rate: %.4f%%\n", switch_error_count, ((double) switch_error_count)/ (condensed_counting.size()-1) * 100);
^~~~~~~~~~~~~~~~~~
hic_switch_error.cpp:595:40: error: unable to find string literal operator 'operator""PRIu64' with 'const char [25]', 'long unsigned int' arguments
printf("Overall total length: %"PRIu64"\n",overall_total_length);
^~~~
hic_switch_error.cpp:596:40: error: unable to find string literal operator 'operator""PRIu64' with 'const char [24]', 'long unsigned int' arguments
printf("\nOverall Unswitched: %"PRIu64"\n", overall_total_vars - overall_hamming_distance);
^~~~
hic_switch_error.cpp:597:36: error: unable to find string literal operator 'operator""PRIu64' with 'const char [21]', 'long unsigned int' arguments
printf("Overall Switched: %"PRIu64"\n", overall_hamming_distance);
^~~~
hic_switch_error.cpp:599:47: error: unable to find string literal operator 'operator""PRIu64' with 'const char [46]', 'long unsigned int' arguments
printf("Overall Switch error number: %"PRIu64", rate: %.4f%%\n", overall_switch_error_count, ((double) overall_switch_error_count)/ overall_potiential_switch_count * 100);
^~~~~~~~~~~~~~~~~~
make: *** [: hic_switch_error.o] Error 1