search

yjzhang / split-seq-pipeline

MIT License
19 stars 21 forks source link

Discrepancies between Readme and split-seq -h, and possibly an unrelated error #8

Closed RobertAlpin closed 5 years ago

RobertAlpin commented 5 years ago

Hi, I'm trying to use your tool and I've noticed that the guide in the Readme seems inconsistent with the -h command inside of the tool itself. The readme suggests that i use --genome-dir (with a dash) while split-seq all seems to expect --genome_dir (with an underscore). I'm also confused about how to to read the .gtf file for running the pipeline. split-seq all doesn't recognize --gtf_file as a valid argument. The -h from the script makes me think --genes is the gtf input I need, but the Readme implies that it's used only for indexing. I've tried using --genes in place of --gtf_file and the pipeline starts, although within about a minute I end up getting the following error:

~~Traceback (most recent call last): File "/extra/robertalpin/pythonpath/bin/split-seq", line 7, in exec(compile(f.read(), file, 'exec')) File "/extra/robertalpin/split-seq-pipeline-master/split-seq", line 79, in valid_samples = analysis.check_valid_samples(args.sample) File "/extra/robertalpin/split-seq-pipeline-master/split_seq/analysis.py", line 186, in check_valid_samples if len(samples)>0: TypeError: object of type 'NoneType' has no len()~~

Is this a result of --genes not being the expected argument? If so, what is the expected argument?

Edit: Did some extra troubleshooting. It seems to run the same way with or without --genes. The error I had seems to be from a lack of --sample input, as it went away when I added some dummy parameters. Is this argument supposed to be optional? It's not included in the Readme example but it gives me the (now struck-through) error above without it.

yjzhang commented 5 years ago

Sorry about that, it seems that the readme was outdated. I've just updated the readme and removed the inaccurate portions. The last example on the page is the correct example for running split-seq all.

The --sample argument shouldn't be required anymore after the latest commit.

As for the .gtf file, currently the pipline requires running split-seq mkref before running split-seq all, and using the output_dir from mkref as the genome_dir in all. The mkref command uses the --genes argument as the .gtf file.

RobertAlpin commented 5 years ago

Thank you so much!