nib-si / pisa-tree Goto Github PK

Missing distinction between wet and dry assay types.
When defining new assay type (e.g. XX) class dry, script states a conflict that the assay type (XX) exits in class wet (and vice versa). At the same the assay type (XX) is not offered (i.e. visible) to choose when creating dry-lab assay (therefore the need for creation of a new type).

I created new Types in Templates/DRY (by choosing Other several times in makeAssay.bat script). When this directory reached 25 subfolders, the makeAssay.bat script exited after selecting DRY and no assay is created (reported by Mojca Juteršek). Deleting a few directories in Templates/DRY resolves the issue.

Option to 'choose' empty type, i.e. no suffix in Assay name

The link on the GitHub landing page for "pISA-tree Web page" (https://nib-si.github.io/pISA/index.html) is broken.

Assay analytes.txt creation:

• i) and ii) two equally valued steps
  i) create assay pISA layer and fill corresponding metadata
  note1: assay has ID as user defined, prefix A as developer defined and suffix -type as user defined
  note2: assay directory name is defined from assay prefix, assay ID and assay sufix
  note3: phenodata is at Investigation level
  ii) mark in the phenodata file which samples are going to be anaylsed in which assay (i.e. preferably prior to assay layer generation, but it can be also conducted post assay layer creation by column addition). column names should correspond to the assay ID-suffix
  iii) additional script run to generate Analytes.txt according to the phenodata column ID--assay ID-suffix cross-link

additional options to considerate
fetch:
assay prefix + assay ID + assay suffix
assay ID + assay suffix
assay ID

The first line in metadata has an unnecessary trailing blank, all layers.

This probably is planned to be changed or the script (Templates/x.lib/upload.bat) removed before release

https://github.com/NIB-SI/pISA/blob/f20fdd956ab856079bb4e5e70b0d7df10b2412ff/Templates/x.lib/upload.bat#L8

Having an Investigation/Study/Assay name in these files makes it impractical to rename the folders after creation. You have to edit all these files or you end up with the old folder name in them. This can lead to confusion.

This issue is most painful at the Assay level where folder moving & renaming is most common.

As the opprtunity to select from various Assay templates, an option to choose from multipe METADATA specifications; e.g.
https://github.com/MIAPPE/MIAPPE/blob/master/MIAPPE_Checklist-Data-Model-v1.1/MIAPPE_mapping/MIAPPE_Checklist-1.1-with-mapping.pdf or other adapted standards

Cross-link funding project information at Investigation metadata level.
Take care not to re-use fields from the common.ini.

When creating a new assay template by choosing WET -> Other, an empty template is generated that includes Keys "Phenodata", "Featuredata" and "Data". However, when you make your own Template these three Keys are not automatically added before the assay type-specific Keys.

Scripts for assas 01_Desc-R

Assas instead of Assay

Control lines for upload and access privileges in Assay METADATA would be preferred;
lines corresponding ones at investigation/Study levels

• Sharing permission
• Upload to FAIRDOMHub

Hi,

I'd like to try this on my mac but it seems to be Windows-specific? I didn't find relevant information in the INSTRUCTIONS.pdf

Thanks a lot, and it seems to be a great tool!

Description: field missing from _PROJECT_METADATA.TXT when using special characters, such as & in description

change the repo name to pISA-tree

Cross-check Assay ID/name/directory and update corresponding filed within belonging: metadata file and all the Read.me's