search

i10labtitech / GINGER

GNU General Public License v3.0
24 stars 2 forks source link

Error in abinitio step #1

Open cappadoc opened 1 year ago

cappadoc commented 1 year ago

Thank you for providing the excellent annotation tool. I used GINGER v1.0.0 (lime-1.0.0) and run each step through nextflow. In the abinitio prediction step (SNAP process) with abinitio.nf, I had an error. The command with the error is the following: python ${script}/down_size.py ${CWD0}/!{augustusTrainTrain} ${CWD0}/!{trainingData} The trainingData is from params.SNAP_TRAINING_DATA, which is ${PDIR_PREP_MAPPING_TOLEARN2ND}/lemon_learn_2nd.gff3, indicated in nextflow.config. However, the lemon_learn_2nd.gff3 file has not been created in the mapping step, but a ginger_learn_2nd.gff3 file been created instead. The execution of the down_size.py script with ginger_learn_2nd.gff3 failed to complete with an error (IndexError: string index out of range). How can I process the SNAP step?

i10labtitech commented 1 year ago

I apologize for the delayed response due to various reasons. Thank you for pointing out the issue. Taking your feedback into account, I've reviewed the code in detail and updated it to version v1.0.1. I apologize for the inconvenience, but could you please clone or download the code and run it again? If you only need to redo the mapping phase, you can use mapping.nf. However, please note that if a directory specified in nextflow.config under PDIR_PREP_MAPPING already exists, the execution will terminate prematurely. So, please remove that directory before running it again.

cappadoc commented 1 year ago

Thank you for your response. Unfortunately, I failed to run the docker image (v1.0.1) because I have no root privileges (a converted singularity image must still have /root access within the container). So, I have installed ginger-1.0.0-0.tar.bz2 obtained at https://anaconda.org/i10labtitech/ginger/files and run through the command B. Could you upload the ginger-1.0.1 version to https://anaconda.org/i10labtitech/ginger/files or to any other site?

Also, the phase0.sh script in ginger-1.0.0 seems to have unclear codes at lines 35 and 36. Is 'output' in these lines correct?

i10labtitech commented 1 year ago

Thank you for letting me know. I will create the v1.0.1 package. I will do my best to complete it as quickly as possible. Creating the conda package for v1.0.1 took quite a bit of time to resolve dependencies, so it might take a little longer. However, I will do my best to finish it as soon as possible. I will respond separately regarding the phase0.sh in version v1.0.0.

i10labtitech commented 1 year ago

As you pointed out, the portion of phase0.sh in the package currently placed in Anaconda was inappropriate. The phase0.sh that is now on GitHub is designed differently. I sincerely apologize for the inconvenience, but could you please wait for the conda package version v1.0.1? I am currently building it, but it seems that resolving dependencies is taking longer than expected.

i10labtitech commented 1 year ago

I apologize for the delay. The issue has been fixed. I created a conda package using boa in an environment built with mamba.

As mentioned in the README, when you run "gingerConfigIni" after installing GINGER with conda or mamba commands, a nextflow.config file will be generated in the current working directory. This file contains the paths to the tools installed simultaneously.

cappadoc commented 1 year ago

Thank you so much. I will try to run the updated ginger, and will report the result.

cappadoc commented 1 year ago

I have successfully completed the run of ginger. The number (~20,000) of the final output proteins (ginger.pep) was reasonable.

However, I have had to re-install and fix several required tools to complete the run. My manipulations added for ginger installation are as follows: (1) the nextflow version installed by conda is v23.04, which is incompatible to the provided nf scripts. So I installed an earlier version of nextflow. (2) gffread, spaln/makeidx.pl, spaln/makblk.pl scripts have not installed by conda. I manually installed these scripts. (3) The installed spaln v3.0.1 was failed to run the command. I re-installed spaln v3.0.0. (4) The installed salmon v0.8.1 was incompatible to the salmon_runner.pl script in Trinity. I installed salmon v1.10.0. (5) The perl path in the 1st lines of GMAP-associated scripts (fa_coords, gmap_process, gmap_cat, and gmap_build.bak) was fixed.

Thank you for your help.

i10labtitech commented 1 year ago

Thank you again for the bug report. I conducted package construction and testing in another environment, but it seems that it was incomplete. To ensure it becomes complete, I plan to prepare a completely new environment and package it.

cappadoc commented 1 year ago

Thank you for your response. I found another serious problems in the final output fasta files (ginger.cds and ginger.pep). It seems that CDSs are incorrectly extracted from transcripts with minus strand ginger_phase2.gff. In addition, several CDSs with plus strand in ginger_phase2.gff do not start with ATG although the corresponding translated proteins start with M. It would be helpful for us to fix these bugs.