Orthomcl running problem

[zillur]

Hi there,
I was trying to run orthomcl in my linux workstation. I am facing this problem:

[root@genomics bin]# ./orthomcl-pipeline -i /home/zillur/Desktop/zillur/phd/orthomcl -o /home/zillur/Desktop/zillur/phd/orthomcl/output -m /usr/local/Tools/orthomcl/orthomcl-pipeline-master/etc/orthomcl.config.example --nocompliant
Warning: directory "/home/zillur/Desktop/zillur/phd/orthomcl/output" already exists, are you sure you want to store data here [Y]? y
Starting OrthoMCL pipeline on: Mon Sep 26 20:11:08 2016
Git commit: unknown

=Stage 1: Validate Files =
Odd number of elements in hash assignment at /root/perl5/lib/perl5/Bio/SeqIO.pm line 378.
Validating PlasmoDB-28_PbergheiANKA_AnnotatedProteins.fasta ... 5076 sequences
Odd number of elements in hash assignment at /root/perl5/lib/perl5/Bio/SeqIO.pm line 378.
Validating PlasmoDB-28_Pchabaudichabaudi_AnnotatedProteins.fasta ... 5217 sequences
Odd number of elements in hash assignment at /root/perl5/lib/perl5/Bio/SeqIO.pm line 378.
Validating PlasmoDB-28_Pfalciparum3D7_AnnotatedProteins.fasta ... 5542 sequences
Odd number of elements in hash assignment at /root/perl5/lib/perl5/Bio/SeqIO.pm line 378.
Validating PlasmoDB-28_Pgallinaceum8A_AnnotatedProteins.fasta ... 3 sequences
Odd number of elements in hash assignment at /root/perl5/lib/perl5/Bio/SeqIO.pm line 378.
Validating PlasmoDB-28_PknowlesiH_AnnotatedProteins.fasta ... 5323 sequences
Odd number of elements in hash assignment at /root/perl5/lib/perl5/Bio/SeqIO.pm line 378.
Validating PlasmoDB-28_PvivaxSal1_AnnotatedProteins.fasta ... 5586 sequences
Odd number of elements in hash assignment at /root/perl5/lib/perl5/Bio/SeqIO.pm line 378.
Validating PlasmoDB-28_PyoeliiyoeliiYM_AnnotatedProteins.fasta ... 5709 sequences
Validated 7 files
Stage 1 took 0.02 minutes 

=Stage 2: Validate Database=
Stage 2 took 0.00 minutes 


=Stage 3: Load OrthoMCL Database Schema=
/usr/bin/orthomclInstallSchema "/usr/local/Tools/orthomcl/orthomcl-pipeline-master/etc/orthomcl.config.example" "/home/zillur/Desktop/zillur/phd/orthomcl/output/log/orthomclSchema.log" 1>/home/zillur/Desktop/zillur/phd/orthomcl/output/log/3.loadschema.stdout.log 2>/home/zillur/Desktop/zillur/phd/orthomcl/output/log/3.loadschema.stderr.log
Error executing command: /usr/bin/orthomclInstallSchema "/usr/local/Tools/orthomcl/orthomcl-pipeline-master/etc/orthomcl.config.example" "/home/zillur/Desktop/zillur/phd/orthomcl/output/log/orthomclSchema.log" 1>/home/zillur/Desktop/zillur/phd/orthomcl/output/log/3.loadschema.stdout.log 2>/home/zillur/Desktop/zillur/phd/orthomcl/output/log/3.loadschema.stderr.log. See logs /home/zillur/Desktop/zillur/phd/orthomcl/output/log/3.loadschema.stdout.log and /home/zillur/Desktop/zillur/phd/orthomcl/output/log/3.loadschema.stderr.log

The log is as follows:

[zillur@genomics log]$ more 3.loadschema.stderr.log 
Can't locate OrthoMCLEngine/Main/Base.pm in @INC (@INC contains: /usr/bin/../lib/perl /root/perl5/lib/perl5/x86_64-linux-thread-multi /root/perl5/lib/perl5 /h
ome/zillur/perl5/lib/perl5/x86_64-linux-thread-multi /home/zillur/perl5/lib/perl5 /usr/local/lib64/perl5 /usr/local/share/perl5 /usr/lib64/perl5/vendor_perl /
usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5 .) at /usr/bin/orthomclInstallSchema line 6.
BEGIN failed--compilation aborted at /usr/bin/orthomclInstallSchema line 6.

Any suggestions please.

Best Regards
Zillur

[zillur]

Is there anybody has any idea? Please. I appreciate your helps.

Best Regards
Zillur

[d_emms]

Hi

It looks like you need to set your PERL5LIB environment variable so that it points to where your orthomcl perl files are. Something like this:
export PERL5LIB=/path/to/orthomcl.

One suggestion though, have you tried OrthoFinder? It's far easier to run, it just requires a single command. It's also a lot more accurate than OrthoMCL:

https://github.com/davidemms/OrthoFinder/releases

David

[zillur]

Thank you very much for your suggestions. Yeah. I have tried orthofinder and it gave me outputs. I wanted to run orthomcl to compare, maybe its not necessary now. Do you have any suggestions how can I process the outputs to get a gene presence/absence matrix?

Thank you again.

Best Regards
Zillur

[d_emms]

The file Orthogroups.csv is effectively a presence/absence matrix: The rows are orthogroups and the columns are species so if there are any genes listed in the i,j-th cell then the ith orthogroup is present in the jth species.

All the best
David

[zillur]

Thank you very much for your comment. I want a matrix like:

              genome1	genome2 genome3
gene1  	 1     	 0     	 0
gene2  	 0     	 0     	 0
gene3  	 1     	 1     	 1
gene4  	 0     	 0     	 1

How can I do this?

Best Regards
Zillur

[d_emms]

You'd just need to replace empty cells with 0 and cells with text in with 1.

All the best
David

[zillur]

Thank you very much for your reply.

You'd just need to replace empty cells with 0 and cells with text in with 1.

Exactly I want to do this. But how can replace this?

Thanks for your suggestions.
Best Regards
Zillur

[d_emms]

This is a python script that will do it for you:

import sys
import csv

if len(sys.argv) != 2:
    print("Usage: python presence_absence.py Orthogroups.csv")
    sys.exit()

inFN = sys.argv[1]
outFN = inFN + ".01_matrix.csv"
with open(inFN, 'rb') as infile, open(outFN, 'wb') as outfile:
    reader = csv.reader(infile, delimiter="\t")
    writer = csv.writer(outfile, delimiter="\t")
    writer.writerow(reader.next())
    for line in reader:
        writer.writerow(line[:1] + [0 if "" == cell else 1 for cell in line[1:]])

All the best
David

[zillur]

Thank you very much for your script. I was trying to run, but:

[zillur@genomics Results_Sep26]$ python matrix_convert_binary.py Orthogroups.csv
Traceback (most recent call last):
  File "matrix_convert_binary.py", line 14, in <module>
    writer.writerow(reader.next())
AttributeError: '_csv.reader' object has no attribute 'next'

My system is:

[zillur@genomics Results_Sep26]$ python
Python 3.5.2 |Continuum Analytics, Inc.| (default, Jul  2 2016, 17:53:06) 
[GCC 4.4.7 20120313 (Red Hat 4.4.7-1)] on linux
Type "help", "copyright", "credits" or "license" for more information.

I am not sure what I need to modify. Any idea?
Thanks again.

Best Regards
Zillur

[d_emms]

It was written for python 2, below is a version which will work with both python 2 and 3:

import sys
import csv

if len(sys.argv) != 2:
    print("Usage: python presence_absence.py Orthogroups.csv")
    sys.exit()

inFN = sys.argv[1]
outFN = inFN + ".01_matrix.csv"
with open(inFN, 'r') as infile, open(outFN, 'w') as outfile:
    reader = csv.reader(infile, delimiter="\t")
    writer = csv.writer(outfile, delimiter="\t")
    writer.writerow(next(reader))
    for line in reader:
        writer.writerow(line[:1] + [0 if "" == cell else 1 for cell in line[1:]])

[zillur]

Thank you very much for your valuable suggestions. The code perfectly converted the matrix into a binary matrix. But the problem is I can't load the new csv file in R as it is:

[zillur@genomics Results_Sep26]$ head Orthogroups.csv.01_matrix.csv 
	PlasmoDB-28_PbergheiANKA_AnnotatedProteins.fasta	PlasmoDB-28_Pchabaudichabaudi_AnnotatedProteins.fasta	PlasmoDB-28_Pfalciparum3D7_AnnotatedProteins.fasta	PlasmoDB-28_Pgallinaceum8A_AnnotatedProteins.fasta	PlasmoDB-28_PknowlesiH_AnnotatedProteins.fasta	PlasmoDB-28_PvivaxSal1_AnnotatedProteins.fasta	PlasmoDB-28_PyoeliiyoeliiYM_AnnotatedProteins.fasta
OG0000000	1	1	0	0	0	0	1
OG0000001	1	1	1	0	1	1	1
OG0000002	0	0	0	0	0	1	0
OG0000003	0	0	0	0	1	1	0
OG0000004	1	1	0	0	0	0	1
OG0000005	0	0	0	0	1	0	0
OG0000006	1	1	0	0	0	0	1
OG0000007	1	1	1	0	1	1	1
OG0000008	0	0	1	0	0	0	0

But when I load the csv in R, it looks like:

> data = read.csv("Orthogroups.csv.01_matrix.csv", sep=",")
> head(data)
  PlasmoDB.28_PbergheiANKA_AnnotatedProteins.fasta.PlasmoDB.28_Pchabaudichabaudi_AnnotatedProteins.fasta.PlasmoDB.28_Pfalciparum3D7_AnnotatedProteins.fasta.PlasmoDB.28_Pgallinaceum8A_AnnotatedProteins.fasta.PlasmoDB.28_PknowlesiH_AnnotatedProteins.fast ...
1                                                                                                                                                                                                                                 OG0000000\t1\t1\t0\t0\t0\t0\t1
2                                                                                                                                                                                                                                 OG0000001\t1\t1\t1\t0\t1\t1\t1
3                                                                                                                                                                                                                                 OG0000002\t0\t0\t0\t0\t0\t1\t0
4                                                                                                                                                                                                                                 OG0000003\t0\t0\t0\t0\t1\t1\t0
5                                                                                                                                                                                                                                 OG0000004\t1\t1\t0\t0\t0\t0\t1
6                                                                                                                                                                                                                                 OG0000005\t0\t0\t0\t0\t1\t0\t0

What should I do now?
Thanks again for your help and comment.

Best Regards
Zillur

[d_emms]

It's a tab-delimited file, try this instead:
data = read.csv("Orthogroups.csv.01_matrix.csv", sep="\t")

[zillur]

Thank you very much. Got it.

Best Regards
Zillur