Polysolver troubleshooting installation <--Help!

I have all the dependencies installed for polysolver.
However, I still get error messages such as:

1) Error: Unable to access jarfile /home/user/Programs//SamToFastq.jar

2) Can't locate Bio/DB/Sam.pm in @INC (you may need to install the Bio::DB::Sam module

For the first error message::::

Picard no longer uses SamToFastq.jar, it contains a SamToFastq.java file instead. Samtools also does not provide a SamToFastq jar, therefore,
I am puzzled as to how I am supposed to make polysolver work if I am unable to find the SamToFastq.jar.

Does anyone know where I can find the SamToFastq jar file?

For the second message::::

Using cpan, I am unable to successfully install the Bio::DB::Sam module. I can manually create the htslib folder and copy paste files from github and that are located elsewhere in the samtools program folder, but there still exists errors that go beyond a simple fix.

I tried to download and manually install Bio-SamTools-1.43 but run into the same problems that I did when using cpan.

Can the polysolver developer fix these issues and make it easier for users to get through the installation process?

Polysolver installation

H SIImlinan ,

I just installed polysolver yeserday and I hit the SamToFastq.jar issue, too. I grabbed picard-tools 1.120 from github and put the path to that in the GATK_DIR line in the config.bash and that got us to the next hurdle. It looks like anything up to and including 1.123 has the tools as individual jars.

When running the tests, I also found that the new samtools has different command line parameters so I put in samtools-0.1.19 and again, pointed SAMTOOLS_DIR to that in the config.bash.

Next, I found that, although the MUTECT_DIR is a variable that you can set, the jar that it looks for is hard coded to muTect-1.1.6.jar on line 292 of scripts/shell_call_hla_mutations_from_type and ours was 1.1.4, so I changed that line.

Also (earlier in the process) ,since our perl is not in /usr/bin, although I installed all of the listed required perl packages with CPAN, I still ran in to trouble finding them. I found that the perl scripts has hashbangs of #!/usr/bin/perl so, even though we had a different one in our path, it didn't pick up it's installed packages. So we had to choose between updating our PERL5LIB. I find some folks at our site reset that variable so changing the hashbangs to "#!/usr/bin/env perl" in scripts/*.pl means that, even if users set their own PERL5LIB, they'll still pick up whichever packages are in the perl install folder of whatever perl they are running. Also, Array::Util was missing in the list of required packages, but we just CPAN installed that too and it's now all running.

I'm not sure which perl you're using or what issues you have when installing Bio-SamTools-1.43. We already had this in so I didn't have to wrestle with it.

Hopefully the tips above can get you a bit further along in the install. We're still ironing out some wrinkles, as the tests in section 3 of the install docs still don't pass so please vet my suggestions before taking them as gospel. Still, hope they help.

Cheers
CanWood

Test 3.2) POLYSOLVER-based mutation detection

UPDATE

After a line by line check and some changes to the scripts I got Polysolever up and running correctly.

Dear CanWood,

Thanks for sharing all this very useful information.

By following your suggestions and changing line 8 of the scripts/merge_fastq.pl file to the path where my x86_64-linux is I managed to pass the HLA typing test 3.1.

I was wondering if you already managed to pass test 3.2 for the mutation calling. I changed the --config path/to/strelka_config_eland_default.ini in line 304 of scripts/shell_call_hla_mutations_from_type and I do not get any errors, HOWEVER the out file does not have any mutations detected (test/call_stats.hla_a_24_02_01_01.out), as the original one has (test/orig.call_stats.hla_b_39_01_01_02l.out).

If I write:

diff test/call_stats.hla_b_39_01_01_02l.out test/orig.call_stats.hla_b_39_01_01_02l.out

I get:


1c1
< ##MuTect:1.1.6-4-g69b7a37 Gatk:3.1-0-g72492bb
---
> ##MuTect:1.1.6-0-g6fe4f4c Gatk:2.7-1-g42d771f
2a3
> hla_b_39_01_01_02l 360 TGAxTGC G C foo foo 0 NOVEL UNCOVERED 0.139341 0.139832 0.99649 0.99649 0 27 0 25.927275 26.000014 8.053996 21.970977 0.583333 0.02 -0.105279 12 5 7 181 262 70 70 0 0 GG 4.513436 0 15 15 0 522 0 0.999725 0.85511 (0,5,2,5) 45 17 30 17 0 KEEP

I wonder if the problem comes from the MuTect and Gatk versions, but if so, I don't know where it is getting those since we don't have installed MuTect-1.1.6 neither GATK-3.1-0.

I would appreciate any help.

Best wishes,
Alejandro

Polysolver installation

Dear CanWood,

I just wanted to thank you for the valuable comments you left here. It helped me to solve my problems much quicker.

Best
tmn