Open paulzierep opened 1 year ago
When trying MaAsLin 2 I found some issues, that make the wrapper basically unusable. (a) The fixed_effects can be different for every dataset, they are however hard-coded to only allow a choice of the test data: https://github.com/galaxyproject/tools-iuc/blob/bc3831e3648809bd3d46611b2a74bd26a17985e5/tools/maaslin2/maaslin2.xml#L57-L64 (b) The correction method should be a choice not text: https://github.com/galaxyproject/tools-iuc/blob/bc3831e3648809bd3d46611b2a74bd26a17985e5/tools/maaslin2/maaslin2.xml#L89 (c) labels for normalization, transform and analysis_method are cryptic (d) the features output would also be nice to have
ping @EngyNasr -> if you find the time, otherwise I or a new bachelor can also look into it when I am back
@paulzierep still valid?
When trying MaAsLin 2 I found some issues, that make the wrapper basically unusable. (a) The fixed_effects can be different for every dataset, they are however hard-coded to only allow a choice of the test data: https://github.com/galaxyproject/tools-iuc/blob/bc3831e3648809bd3d46611b2a74bd26a17985e5/tools/maaslin2/maaslin2.xml#L57-L64 (b) The correction method should be a choice not text: https://github.com/galaxyproject/tools-iuc/blob/bc3831e3648809bd3d46611b2a74bd26a17985e5/tools/maaslin2/maaslin2.xml#L89 (c) labels for normalization, transform and analysis_method are cryptic (d) the features output would also be nice to have