Closed lichen-fungus closed 1 year ago
That's weird.
What is the value of $installpath
in the /data/scratch/projects/p_citrina_transcriptome/SqueezeMeta_conf.pl
file?
Yes weird. Unfortunately, $installpath
is not defined in the SqueezeMeta_conf.pl
, I see only$databasepath
, $extdatapath = "$installpath/data"
and $scriptdir = "$installpath/scripts"
defined in SqueezeMeta_conf.pl
in the section #-- Generic paths
.
I luckily found the place where the $installpath
defined: it is in SqueezeMeta.pl
at the very top of the file.
###scriptdir patch v2, Fernando Puente-Sánchez, 18-XI-2019
use File::Basename;
use Cwd 'abs_path';
our $scriptdir;
if(-l __FILE__)
{
my $symlinkpath = dirname(__FILE__);
my $symlinkdest = readlink(__FILE__);
$scriptdir = dirname(abs_path("$symlinkpath/$symlinkdest"));
}
else
{
$scriptdir = abs_path(dirname(__FILE__));
}
our $installpath = abs_path("$scriptdir/..");
###
Do you think it may work if I use the following fix in SqueezeMeta.pl
, replacing the code above by the code below? I hope $symlinkpath
or $symlinkdest
or FILE
are not used elsewhere.
###my own scriptdir patch
# use File::Basename;
use Cwd 'abs_path';
our $scriptdir = "~/anaconda3/envs/SqueezeMeta/SqueezeMeta/scripts/";
our $installpath = abs_path("$scriptdir/..");
###
Note that I mentioned the SqueezeMeta_conf.pl
file inside the project directory. That one should have $installpath
defined.
E.g.
1) I just ran the command SqueezeMeta.pl -m sequential -f raw -s test2.samples
2) This created a directory named H1
(the name of one of my samples)
3) This are the first lines of the H1/SqueezeMeta_conf.pl
file
$version = "1.6.2, March 2023";
$mode = "sequential";
$date = "Fri Jul 21 16:38:01 2023";
$installpath = "/home/fer/.local/opt/miniconda3/envs/SqueezeMeta/SqueezeMeta";
$userdir = "/data/fer/SQMtestData/Hadza/raw";
$installpath
is defined right at the beginning.
How did you install the databases? By any chance did you rename the file SqueezeMeta_conf_original.pl
inside the SqueezeMeta scripts directory to SqueezeMeta.pl
?
So answering to your previous question, I wouldn't recommend the fix you suggest. If anything, you should instead edit the SqueezeMeta_conf.pl
file inside the project's directory (note that this is not the same as the file inside SqueezeMeta's script directory in the source code) and add the right value for $installpath
at the beginning.
But this should already be present without the need for you to add anything (it works for me using the latest version and I don't remember this having been an issue in the past). So I would rather like to understand what is the origin of your current issue.
Try going below the directory named p-citrina2/
(the one that contains the data
. results
, etc folders
and then restart again with SqueezeMeta.pl -p p-citrina2 --restart -step 2
Sobre 21/07/2023 17:50:45, lichen-fungus @.***> escribió:
Thanks a lot for all your help!
I don't remember renaming any files inside the SqueezeMeta scripts directory, and my SqueezeMeta.pl there looks different from the config file. (By the way SqueezeMeta at least doesn't die on the first step when used with my fix but the original fix is safer so I removed my fix again.) The trouble is, if I start from scratch, there is no modified SqueezeMeta_config.pl because there is not yet a project folder.
I got the databases by downloading them with the download script download_databases.pl. I then symlinked them onto a folder that belongs to root, afterwards removed the symlink because it was throwing errors when I tried to use the SqueezeMeta databases in that location. Maybe that's where things somehow got messed up. — Reply to this email directly, view it on GitHub [https://github.com/jtamames/SqueezeMeta/issues/715#issuecomment-1645791774], or unsubscribe [https://github.com/notifications/unsubscribe-auth/ACAHO7XUPDOTP27QD3E3U43XRKQNJANCNFSM6AAAAAA2S3ZYSY]. You are receiving this because you commented.Message ID: @.***> [b0aa5bdc-0a7a-4168-ad4e-5172736c90a5]
Hi Fernando,
I tried to restart the older analysis that ran until step 16 from one level above the p-citrina2 directory containing the data/
and results/
after copying the samples file into the p-citrina2
folder and it starts and dies: SqueezeMeta.pl -p p-citrina2 --restart -step 14
SqueezeMeta v1.6.2, March 2023 - (c) J. Tamames, F. Puente-Sánchez CNB-CSIC, Madrid, SPAIN
Please cite: Tamames & Puente-Sanchez, Frontiers in Microbiology 9, 3349 (2019). doi: https://doi.org/10.3389/fmicb.2018.03349
Run started Fri Jul 21 18:28:27 2023 in sequential mode
1 metagenomes found: p-citrina2
--- SAMPLE p-citrina2 ---
Binning results for concoct,metabat2 already found, skipping step 14
DASTool results in /data/scratch/projects/p_citrina_transcriptome/tmp/SqueezeMeta/restart_old/p-citrina2/results/bins already found, skipping step 15
DASTool results in /data/scratch/projects/p_citrina_transcriptome/tmp/SqueezeMeta/restart_old/p-citrina2/intermediate/16.p-citrina2.bintax already found, skipping step 16
[0 seconds]: STEP17 -> CHECKING BINS: 17.checkM_batch.pl
Can't find SqueezeMeta_conf.pl in /data/scratch/projects/p_citrina_transcriptome/tmp/SqueezeMeta/restart_old/p-citrina2/p-citrina2. Is the project path ok? at /home/admin/anaconda3/envs/SqueezeMeta/SqueezeMeta/scripts/17.checkM_batch.pl line 17.
Stopping in STEP17 -> 17.checkM_batch.pl
Died at /home/admin/anaconda3/envs/SqueezeMeta/bin/SqueezeMeta.pl line 816.
If I move into the p-citrina2
directory containing data/
and results/
and execute the same code, then I get this error message:
***
Can't open samples file (-s) in p-citrina2/data/00..samples. Please check that it is the correct file
in SqueezeMeta_conf.pl inside the p-citrina2 directory I set the project directory like this because with the original code it wouldn't work:
$projectdir = "/data/scratch/projects/p_citrina_transcriptome/tmp/SqueezeMeta/restart_old/p-citrina2";
Sorry to bother you with this problem but no matter what I do there is still something not quite right...
Hi again,
What is the content of the /data/scratch/projects/p_citrina_transcriptome/tmp/SqueezeMeta/restart_old/p-citrina2/p-citrina2
directory?
I suspect that you'd be better reinstalling from scratch following the instructions. It seems that something in your install got broken along the way...
Closing due to lack of activity, feel free to reopen
when running SqueezeMeta in Sequential mode like this:
SqueezeMeta.pl -m sequential -s samples -f /data/scratch/ projects/p_citrina_transcriptome/tmp/trimmomatic/ -extassembly /data/scratch/ projects/p_citrina_transcriptome/tmp/trinity/P_citrina_metatranscriptome_unfiltered_Trinity.fasta
, I get the following error message:The variable
$installpath
is wrongly defined adding/data
in front of the correct path, and therefore thejgi_summarize_bam_contig_depths
script can't be found. Can you please advise how I can change$installpath
to the correct path, or where $installpath gets defined? I would prefer not to have to change all instances of$installpath
in the configuration fileSqueezeMeta_conf.pl
if that can be avoided somehow.