how to set paramenters without cluster?

[mili-ai]

I don't have a cluster on my server, only regular CPUs. How should I set thread parameters?

[MichaelHiller]

Oh, good question. This can take a long time.
@kirilenkobm Does TOGA have the ability to run multi-threaded?

You might want to wait a few more weeks until we release TOGA 2.0 which will be much much faster and >300X less memory demanding.

[mili-ai]

Thank you for your reply. I am looking forward to TOGA 2.0 version

[ReverendCasy]

Hello @mili-ai ,
Sorry for the late response. In the current TOGA versions, there are two ways to run it on your local machine:

1. Do not set a --nextflow_config_dir option. This will automatically set Nextflow executor to ‘local’.
2. Modify Nextflow configuration files in the nextflow_config_files/ directory by setting process.executor to local, then provide the path to TOGA with a --nextflow_config_dir option. This way also allows you to control the number of CPUs used by each paralleled TOGA step.
   Note that some of CESAR jobs might take a lot of memory to compute, so consider setting the upper memory threshold with --cesar_mem_limit.
   Hope that helps. Please let me know if you tried any of these options and whether they worked fine for you.

Best,
Yury

[swagttt1]

Oh, good question. This can take a long time. @kirilenkobm Does TOGA have the ability to run multi-threaded?

You might want to wait a few more weeks until we release TOGA 2.0 which will be much much faster and >300X less memory demanding.

Thank you for developing such a useful tool! It is currently December 12, 2024. When will TOGA v2.0 be released?

[MichaelHiller]

There is no multithreaded option. We parallelize over the transcripts.

TOGA2 will be released in early January.

[swagttt1]

Hello @mili-ai , Sorry for the late response. In the current TOGA versions, there are two ways to run it on your local machine:

1. Do not set a option. This will automatically set Nextflow executor to ‘local’.--nextflow_config_dir
2. Modify Nextflow configuration files in the directory by setting to , then provide the path to TOGA with a option. This way also allows you to control the number of CPUs used by each paralleled TOGA step.
   Note that some of CESAR jobs might take a lot of memory to compute, so consider setting the upper memory threshold with .
   Hope that helps. Please let me know if you tried any of these options and whether they worked fine for you.nextflow_config_files/``process.executor``local``--nextflow_config_dir``--cesar_mem_limit

Best, Yury

Hi, Yury
Hope this message finds you well!
I don't have a Slurm cluster and have been running TOGA on a local CentOS server. Based on discussions in another issue (#131), I found that TOGA may not be compatible with a single machine and is better suited for running on a Slurm cluster. In this issue, following your guidance, I ran the command:
./toga.py test_input/hg38.mm10.chr11.chain test_input/hg38.genCode27.chr11.bed test_input/hg38.2bit test_input/mm10.2bit --kt --pn test -i supply/hg38.wgEncodeGencodeCompV34.isoforms.txt --cb 3,5 --cjn 500 --u12 supply/hg38.U12sites.tsv --ms
(removing the --nc nextflow_config_files parameters), but encountered a new problem:

......
chain_runner_55: processing chain_id: 553928 transcripts: ENST00000320048,
chain_runner_55: processing chain_id: 513374 transcripts: ENST00000329434,ENST00000328375,ENST00000641167,
chain_runner_55: processing chain_id: 749433 transcripts: ENST00000301790,ENST00000540857,ENST00000539221,
chain_runner_55: processing chain_id: 86290 transcripts: ENST00000313555,ENST00000641320,
chain_runner_55: processing chain_id: 930267 transcripts: ENST00000625203,
chain_runner_55: processing chain_id: 653133 transcripts: ENST00000328188,

#### STEP 3: Merge step 2 output

Reading /home/liuyt/software/TOGA/test/temp/toga_filt_ref_annot.bed
merge_chains_output: got data for 3674 transcripts
merge_chains_output: Loading the results...
merge_chains_output: There are 60 result files to combine
merge_chains_output: got 17036 keys in chain_genes_data
merge_chains_output: got 20495 keys in chain_raw_data
merge_chains_output: There were 20482 transcript lines and 20495 chain lines
merge_chains_output: ERROR!
genes_counter and chain_counter hold different values:
20482 and 20495 respectively

Do you have any insights or suggestions?
Thanks

[ReverendCasy]

Hi @swagttt1 ,
I guess some of the feature extraction jobs exited improperly. Could you please attach the whole TOGA log and/or feature extraction Nextflow log?

Best,
Yury