Metabolic-c test files stuck at Prodigal

[lacy111693]

I have downloaded all the dependencies and was able to run metabolic-g on the test files successfully. However, when trying to run metabolic-c on the test files, it gets stuck on the prodigal step. I have tried several times, and this is where it always gets stuck. I usually close it after a couple of days then try again. I decided to leave it be this time and see if it would ever move, and it's currently been at the prodigal step for 5 days. It's just the test files, so I would not think the size would be an issue.

I am using metabolic within Anaconda Navigator if that helps, and on a Mac with Sequoia 15.1.1, the platform is osx-64.

Note in the screenshot, it says that some of the directories can't be created because they already exist since I have already tried this several times.

[lacy111693]

@patriciatran

[patriciatran]

Hi @lacy111693 ,

I don't work on maintaining this software, but was able to find the mkdir problem so I fixed it.
I added the -p flag in the mkdir commands such that the program will only make a directory if the folder doesn't exist already.

As for the second question about prodigal, prodigal is not multithreaded so it can take a long time. If I were you, I would use something like pprodigal https://github.com/sjaenick/pprodigal and use that for metabolic-g. That said metabolic-c will take unannotated genomes only, so that is not of help here.

It is strange that it would take over 5 days to run the test file with 5 genomes though. I would check the Anaconda navigator set-up, and see if you are running into any disk, memory, or thread issues.

[lacy111693]

Thanks @patriciatran, I discovered that the error is actually GTDB-Tk halting at the prodigal steps. When trying to run and debug GTDB-Tk alone, I got these errors.

[2025-02-07 13:42:50] INFO: GTDB-Tk v2.4.0
[2025-02-07 13:42:50] INFO: gtdbtk classify_wf --cpus 1 -x fasta --genome_dir /Users/lacy_barrett/METABOLIC/METABOLIC_test_files/Guaymas_Basin_genome_files --skip_ani_screen --out_dir test_gtdbtk --debug
[2025-02-07 13:42:50] INFO: Using GTDB-Tk reference data version r220: /Users/lacy_barrett/.conda/envs/metabolic/share/gtdbtk-2.4.0/db
[2025-02-07 13:42:50] INFO: Identifying markers in 5 genomes with 1 threads.
[2025-02-07 13:42:50] TASK: Running Prodigal V2.6.3 to identify genes.
==> Processed 0/5 genomes (0%) | | [?genome/s, ETA ?]Traceback (most recent call last):
File "", line 1, in
File "/Users/lacy_barrett/.conda/envs/metabolic/lib/python3.10/multiprocessing/spawn.py", line 116, in spawn_main
exitcode = _main(fd, parent_sentinel)
File "/Users/lacy_barrett/.conda/envs/metabolic/lib/python3.10/multiprocessing/spawn.py", line 126, in _main
self = reduction.pickle.load(from_parent)
AttributeError: 'Parallel' object has no attribute '__process_manager'

When I went to the GTDB-Tk github page to look into issues, other people seemed to have a similar problem, and the response was always that GTDB-Tk does not work on mac because of pplacer. But metabolic says it works on mac, even though GTDB-Tk is one of the dependencies. So I'm trying to figure out how that works?

[patriciatran]

Hello @lacy111693 , can you link me to where you saw that METABOLIC works on Mac? I have personally never tried it on Mac, but I can fix any documentation that says otherwise.

[lacy111693]

Hello @lacy111693 , can you link me to where you saw that METABOLIC works on Mac? I have personally never tried it on Mac, but I can fix any documentation that says otherwise.

Let me see if I can find it and I'll link it in a comment! I may be misremembering, it was over a year ago that I started digging into metabolic and thought I had read that it works on mac. To be fair though, I'm not a computer person, so it's entirely possible when I started looking through dependencies and saw that some would not work on windows, then I assumed mac was the only option. I did not know what linux was at the time.

[lacy111693]

@patriciatran I was able to run linux as a subsystem on my windows computer and got everything downloaded, woo! When running the test files, I did still run into a few issues. A readline closure at 1908 and issues with the metabolic handoff diagrams as well as the energy plots. I just ran perl METABOLIC-C.pl -test true with no changes to any of the code. Any thoughts? Thanks so much for all of your help.

[2025-02-18 12:31:54] The Prodigal annotation is running...
[2025-02-18 12:32:31] The Prodigal annotation is finished
readline() on closed filehandle _IN at METABOLIC-C.pl line 1908.
readline() on closed filehandle _IN at METABOLIC-C.pl line 1908.
[2025-02-18 12:32:32] The hmmsearch is running with 20 cpu threads...
[2025-02-18 12:57:24] The hmmsearch is finished
[2025-02-18 12:57:39] Generating each hmm faa collection...
[2025-02-18 12:57:39] Each hmm faa collection has been made
[2025-02-18 12:57:39] The KEGG module result is calculating...
[2025-02-18 13:00:12] The KEGG identifier (KO id) result is calculating...
[2025-02-18 13:00:12] The KEGG identifier (KO id) seaching result is finished
[2025-02-18 13:00:12] Searching CAZymes by dbCAN2...
[2025-02-18 13:01:47] dbCAN2 searching is done
[2025-02-18 13:01:47] Searching MEROPS peptidase...
[2025-02-18 13:02:18] MEROPS peptidase searching is done
[2025-02-18 13:02:19] METABOLIC table has been generated
[2025-02-18 13:02:19] Drawing element cycling diagrams...
[2025-02-18 13:06:51] Drawing element cycling diagrams finished
[2025-02-18 13:06:51] Drawing metabolic handoff diagrams...
mv: cannot stat 'METABOLIC_out/newdir/Bar_plot/bar_plot_input_1.pdf': No such file or directory
mv: cannot stat 'METABOLIC_out/newdir/Bar_plot/bar_plot_input_1.pdf': No such file or directory
mv: cannot stat 'METABOLIC_out/newdir/Bar_plot/bar_plot_input_2.pdf': No such file or directory
mv: cannot stat 'METABOLIC_out/newdir/Bar_plot/bar_plot_input_2.pdf': No such file or directory
[2025-02-18 13:06:52] Drawing metabolic handoff diagrams finished
[2025-02-18 13:06:52] Drawing energy flow chart...
[2025-02-18 13:06:52] INFO: GTDB-Tk v2.4.0
[2025-02-18 13:06:52] INFO: gtdbtk classify_wf --cpus 20 -x fasta --genome_dir /home/lacy111693/METABOLIC/METABOLIC_test_files/Guaymas_Basin_genome_files --skip_ani_screen --out_dir METABOLIC_out/intermediate_files/gtdbtk_Genome_files
[2025-02-18 13:06:52] INFO: Using GTDB-Tk reference data version r220: /home/lacy111693/anaconda3/envs/metabolic/share/gtdbtk-2.4.0/db
[2025-02-18 13:06:52] INFO: Identifying markers in 5 genomes with 20 threads.
[2025-02-18 13:06:52] TASK: Running Prodigal V2.6.3 to identify genes.
[2025-02-18 13:07:09] INFO: Completed 5 genomes in 16.60 seconds (3.32 seconds/genome).
[2025-02-18 13:07:09] TASK: Identifying TIGRFAM protein families.
[2025-02-18 13:07:12] INFO: Completed 5 genomes in 3.21 seconds (1.56 genomes/second).
[2025-02-18 13:07:12] TASK: Identifying Pfam protein families.
[2025-02-18 13:07:12] INFO: Completed 5 genomes in 0.25 seconds (20.05 genomes/second).
[2025-02-18 13:07:12] INFO: Annotations done using HMMER 3.4 (Aug 2023).
[2025-02-18 13:07:12] TASK: Summarising identified marker genes.
[2025-02-18 13:07:12] INFO: Completed 5 genomes in 0.09 seconds (54.60 genomes/second).
[2025-02-18 13:07:12] INFO: Done.
[2025-02-18 13:07:13] INFO: Aligning markers in 5 genomes with 20 CPUs.
[2025-02-18 13:07:13] INFO: Processing 5 genomes identified as bacterial.
[2025-02-18 13:09:57] INFO: Read concatenated alignment for 107,235 GTDB genomes.
[2025-02-18 13:09:57] TASK: Generating concatenated alignment for each marker.
mv: cannot stat 'METABOLIC_out/Output_energy_flow/Energy_plot/network.plot.pdf': No such file or directory
mv: cannot stat 'METABOLIC_out/Output_energy_flow/Energy_plot/network.plot.pdf': No such file or directory
[2025-02-18 13:10:02] Drawing energy flow chart finished
[2025-02-18 13:10:02] Calculating MW-score ...
[2025-02-18 13:10:03] Calculating MW-score is done
METABOLIC-C was done, the total running time: 00:38:09 (hh:mm:ss)