search

jkimlab / DESCHRAMBLER

21 stars 8 forks source link

Segmentation error #20

Closed Alchimic007 closed 10 months ago

Alchimic007 commented 1 year ago

Hello, I'm really sorry for bothering you again. I've got an error while trying to run my first experiment with DESCHRAMBLER. I'm running it on virtual machine (Ubuntu 18.04). Here is my tree ((Homo_sapiens:0.00644,Bos_taurus:0.0041)@:0.5689,Choloepus_hoffmanni:0.005); Here is my directory structure /home/maksym/Downloads/Deschrambler_test/chainNet/Homo_sapiens/Bos_taurus/chain/chr1.chain ... chr22 and /home/maksym/Downloads/Deschrambler_test/chainNet/Homo_sapiens/Bos_taurus/net/chr1.net and same for Choloepus_hoffmanni directory. Chains and nets were created from cactus hal file https://cgl.gi.ucsc.edu/data/cactus/241-mammalian-2020v2.hal using UCSC (maftopsl, axtChain, chainNet) and HALtools toolkit (hal2maf) apps. When I'm running DESCHRAMBLER it goes like this: maksym@maksym-virtualBox:~$ perl /home/maksym/Downloads/DESCHRAMBLER/DESCHRAMBLER.pl /home/maksym/Downloads/Deschrambler_test/params.txt

Constructing syntenic fragments

Segmentation fault (core dumped)

make: *** [Makefile:57: Genomes.Order] Error 139

TREE ((Homo_sapiens:0.00644,Bos_taurus:0.0041)@:0.5689,Choloepus_hoffmanni:0.005);

0 Homo_sapiens /home/maksym/Downloads/Deschrambler_test/tree.txt /home/maksym/Downloads/Deschrambler_test/APCFs.300/SFs/bpdist.txt

Illegal division by zero at /home/maksym/Downloads/DESCHRAMBLER/script/estparJC.pl line 41, line 1.

Estimate JC parameter:

inferAdjProb - inferring the posterior probability of block adjacency

usage: inferAdjProb refspc parameter-alpha tree-file genome-file

readline() on closed filehandle F at /home/maksym/Downloads/DESCHRAMBLER/script/refine_adjprob.pl line 11.

Minimum weight = 0.0001

Conservation score file = /home/maksym/Downloads/Deschrambler_test/APCFs.300/SFs/block_consscores.txt

Segmentation fault (core dumped)

maksym@maksym-virtualBox:~$

Alchimic007 commented 1 year ago

I also give you my chain and net example to check. If you will have some time please look at them, may be problem is in my inputs chr1.chain.txt chr1.net.txt I'm sending txt version of examples of my chains and nets (bos_taurus in here), because github does not allow originals. Maybe this will help to understand what is a problem, because I do not get a clue what is wrong. Thank you and sorry for bothering

Alchimic007 commented 1 year ago

I have also re-run examples, they works fine

jkimlab commented 1 year ago

Please search the closed issues in Github first. There were some issues related to the segmentation fault.

On Nov 21, 2023, at 12:42 AM, Alchimic007 @.***> wrote:

I have also re-run examples, they works fine

— Reply to this email directly, view it on GitHub https://github.com/jkimlab/DESCHRAMBLER/issues/20#issuecomment-1819310687, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEU6PV3DGPGC3EGAHRTSBFDYFN25VAVCNFSM6AAAAAA7TDMLBWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMJZGMYTANRYG4. You are receiving this because you are subscribed to this thread.

Alchimic007 commented 1 year ago

I have checked all issues concerning segmentation fault error, namely #1, 7, 13, 18, and it seems they are not applying to my case fully, because in my case files are extracted from hal file and not generated from fasta files. As for naming of files as mentioned in issue #7, my files were named that way as they are from the start: maksym@maksym-virtualBox:~/Downloads/Deschrambler_test/chainNet/Homo_sapiens/Bos_taurus/chain$ ls

chr10.chain chr12.chain chr14.chain chr16.chain chr18.chain chr1.chain chr21.chain chr2.chain chr4.chain chr6.chain chr8.chain

chr11.chain chr13.chain chr15.chain chr17.chain chr19.chain chr20.chain chr22.chain chr3.chain chr5.chain chr7.chain chr9.chain

maksym@maksym-virtualBox:~/Downloads/Deschrambler_test/chainNet/Homo_sapiens/Bos_taurus/chain$

I just want to understand wheter I need to regenerate chains and nets or the problem is in parameters here is my parameter and config files: params file# Reference species REFSPC=Homo_sapiens

Output directory

OUTPUTDIR=/home/maksym/Downloads/Deschrambler_test/APCFs.300

OUTPUTDIR=/home/maksym/Downloads/Deschrambler_test/Out_RACA.anctest

Block resolution (bp)

RESOLUTION=300000

Newick tree file

Refer to the sample file 'tree.txt'.

TREEFILE=/home/maksym/Downloads/Deschrambler_test/tree.txt

Minimum adjacency scores

MINADJSCR=0.0001

Config and make files for syntenic fragment construction

Refer to the sample files 'config.SFs' and 'Makefile.SFs'.

You need to change settings in the sample configuration file (config.SFs) according to your data

You don't need to change anything in the Makefile.SFs.

CONFIGSFSFILE=/home/maksym/Downloads/Deschrambler_test/config.SFs MAKESFSFILE=/home/maksym/Downloads/Deschrambler_test/Makefile.SFs

****config file:# path of a net files

netdir /home/maksym/Downloads/Deschrambler_test/chainNet

path of a chain files

chaindir /home/maksym/Downloads/Deschrambler_test/chainNet

species-name tag1 tag2

tag1 (0: ref-species, 1: descendents, 2: outgroup)

tag2 (1: chromosome assembly, 0: others)

Warning for tag2: Use the value 1 only when a chromosome name starts with "chr".

species Homo_sapiens 0 1 Bos_taurus 1 0 Choloepus_hoffmanni 2 0

block resolution (bp): DO NOT CHANGE

resolution

I will be vey appreciated for any help. Thank you
jkimlab commented 1 year ago

I think it may be caused by the incompatibility of the used cactus alignments. I recommend using the lastz alignments for making chain/net files.

On Nov 21, 2023, at 1:08 AM, Alchimic007 @.***> wrote:

I have checked all issues concerning segmentation fault error, namely #1 https://github.com/jkimlab/DESCHRAMBLER/issues/1, 7, 13, 18, and it seems they are not applying to my case fully, because in my case files are extracted from hal file and not generated from fasta files. As for naming of files as mentioned in issue #7 https://github.com/jkimlab/DESCHRAMBLER/issues/7, my files were named that way as they are from the start: @.***:~/Downloads/Deschrambler_test/chainNet/Homo_sapiens/Bos_taurus/chain$ ls

chr10.chain chr12.chain chr14.chain chr16.chain chr18.chain chr1.chain chr21.chain chr2.chain chr4.chain chr6.chain chr8.chain

chr11.chain chr13.chain chr15.chain chr17.chain chr19.chain chr20.chain chr22.chain chr3.chain chr5.chain chr7.chain chr9.chain

@.***:~/Downloads/Deschrambler_test/chainNet/Homo_sapiens/Bos_taurus/chain$

I just want to understand wheter I need to regenerate chains and nets or the problem is in parameters here is my parameter and config files: params file# Reference species REFSPC=Homo_sapiens

Output directory

OUTPUTDIR=/home/maksym/Downloads/Deschrambler_test/APCFs.300

OUTPUTDIR=/home/maksym/Downloads/Deschrambler_test/Out_RACA.anctest

Block resolution (bp)

RESOLUTION=300000

Newick tree file

Refer to the sample file 'tree.txt'.

TREEFILE=/home/maksym/Downloads/Deschrambler_test/tree.txt

Minimum adjacency scores

MINADJSCR=0.0001

Config and make files for syntenic fragment construction

Refer to the sample files 'config.SFs' and 'Makefile.SFs'.

You need to change settings in the sample configuration file (config.SFs) according to your data

You don't need to change anything in the Makefile.SFs.

CONFIGSFSFILE=/home/maksym/Downloads/Deschrambler_test/config.SFs MAKESFSFILE=/home/maksym/Downloads/Deschrambler_test/Makefile.SFs

****config file:# path of a net files

netdir /home/maksym/Downloads/Deschrambler_test/chainNet

path of a chain files

chaindir /home/maksym/Downloads/Deschrambler_test/chainNet

species-name tag1 tag2

tag1 (0: ref-species, 1: descendents, 2: outgroup)

tag2 (1: chromosome assembly, 0: others)

Warning for tag2: Use the value 1 only when a chromosome name starts with "chr".

species Homo_sapiens 0 1 Bos_taurus 1 0 Choloepus_hoffmanni 2 0

block resolution (bp): DO NOT CHANGE

resolution

I will be vey appreciated for any help. Thank you — Reply to this email directly, view it on GitHub https://github.com/jkimlab/DESCHRAMBLER/issues/20#issuecomment-1819357037, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEU6PV2BNLM5CJ3ZD66AVQTYFN56HAVCNFSM6AAAAAA7TDMLBWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMJZGM2TOMBTG4. You are receiving this because you commented.