Experiment Page: Add processing information modals

[ramenhog]

Description

When user clicks on the link in the "Processing Information" column in the samples table, they should get a modal that displays all the processing information associated with that sample.

Mockup

refine.bio processed with 1 middle step

refine.bio processed with 2 middle steps

submitter processed

[ramenhog]

@Miserlou correct me if I'm wrong, but we don't get this info from the samples endpoint yet right?

[arielsvn]

It seems so, @ramenhog. I checked the latest version of the samples endpoint and it's returning this information as a string under annotations

"data_processing": "['Expression data (CEL files) were normalized with the MAS5.0 algorithm (target signal = 100) using GCOS software (Affymetrix).']",

I guess it will need further processing to be able to extract this information. Do we have an issue for this in the backend??

Entire response for reference `/samples/5/`

```json { "id": 5, "title": "Neuroblastoma, #5", "accession_code": "GSM414004", "organism": { "name": "HOMO_SAPIENS", "taxonomy_id": 9606 }, "annotations": [ { "data": { "type": "['RNA']", "title": "['Neuroblastoma, #5']", "status": "['Public on Mar 08 2012']", "label_ch1": "['biotin']", "series_id": "['GSE16254', 'GSE16476']", "taxid_ch1": "['9606']", "description": "['ITCC0009-inss-2']", "platform_id": "['GPL570']", "contact_city": "['Amsterdam']", "contact_name": "['Rogier,,Versteeg']", "hyb_protocol": "['10 µg of labeled cRNA was hybridized to Affymetrix Human genome U133 Plus 2.0 arrays according to manufacturer’s protocol (Affymetrix).']", "molecule_ch1": "['total RNA']", "organism_ch1": "['Homo sapiens']", "channel_count": "['1']", "contact_email": "['r.versteeg@amc.uva.nl']", "geo_accession": "['GSM414004']", "scan_protocol": "['Arrays were scanned with a GeneChip Scanner 3000 (Affymetrix) according to manufacturer’s protocol.']", "data_row_count": "['54675']", "contact_address": "['P.O. Box 22700']", "contact_country": "['Netherlands']", "data_processing": "['Expression data (CEL files) were normalized with the MAS5.0 algorithm (target signal = 100) using GCOS software (Affymetrix).']", "source_name_ch1": "['Neuroblastoma tumor']", "submission_date": "['Jun 07 2009']", "last_update_date": "['Mar 08 2012']", "contact_institute": "['Academic Medical Centre, University of Amsterdam']", "contact_department": "['Department of Oncogenomics']", "label_protocol_ch1": "['4 µg of total RNA was used for cRNA synthesis and fragmented. Labeling was performed with One-Cycle cDNA synthesis Kit (Affymetrix) according to manufacturer’s protocol. Sample quality was checked on a Bioanalyzer prior and after fragmentation.']", "supplementary_file": "['ftp://ftp.ncbi.nlm.nih.gov/geo/samples/GSM414nnn/GSM414004/suppl/GSM414004.CEL.gz']", "characteristics_ch1": "['tissue: Neuroblastoma tumor']", "extract_protocol_ch1": "['Total RNA was isolated using Trizol (Invitrogen) and purified with RNeasy (Qiagen) according to manufacturer’s protocol. RNA sample integrity was checked on an Agilent 2100 Bioanalyzer (Agilent Technologies).']", "contact_zip/postal_code": "['1100DE']" }, "is_ccdl": false, "created_at": "2018-05-30T18:41:29.962438Z", "last_modified": "2018-05-30T18:41:29.962438Z" } ], "results": [], "source_archive_url": "", "has_raw": true, "sex": "", "age": null, "specimen_part": "neuroblastoma tumor", "genotype": "", "disease": "", "disease_stage": "", "cell_line": "", "treatment": "", "race": "", "subject": "", "compound": "", "time": "", "is_downloaded": false, "is_processed": false, "created_at": "2018-05-30T18:41:29.960150Z", "last_modified": "2018-05-30T18:41:29.960150Z" } ```

[cgreene]

~My impression is that this isn't landing until after Keytar Kurt. It looks like it will depend at least on https://github.com/AlexsLemonade/refinebio/issues/101 which is not in the Keytar milestone.~ Edit: edited to account for @jaclyn-taroni's reply. Basically look 👇

[jaclyn-taroni]

I would argue that it requires the copy on AlexsLemonade/refinebio#170

[cgreene]

Oh! Spiffy! Then is https://github.com/AlexsLemonade/refinebio/issues/101 going to be different than what is in https://github.com/AlexsLemonade/refinebio/issues/170?

[arielsvn]

Ok, I'll remove this issue from the Keytar Kurt milestone.

[jaclyn-taroni]

I'm imagining that it will be a bit more comprehensive in that it will include links to repositories where we've done bits of analysis, etc.

[ramenhog]

Ok, just to confirm, we don't need to worry about showing any processing information for Keytar Kurt?

[jaclyn-taroni]

we don't need to worry about showing any processing information for Keytar Kurt?

I think we do, because we have the copy needed. What do you think @cgreene?

[cgreene]

Actually, I got confused by a separate issue that I've been watching in the waffle. It looks like there is enough information in https://github.com/AlexsLemonade/refinebio/issues/170 to do a first iteration of this page.

[jaclyn-taroni]

I believe this is also related (package versions, etc): AlexsLemonade/refinebio#229

[arielsvn]

I read through these issues, but I still have one question: How can I know which pipeline was used to process a given sample??

@jaclyn-taroni told me the field is_ccdl can be used to identify when the Sample was processed by the submitter.

Also, we need to show version information for these tools: Salmon, Scan, txtimport, etc https://github.com/AlexsLemonade/refinebio/issues/229 talks about this... Is this information exposed on any endpoint??

[arielsvn]

Tagging @Miserlou, we just spoke about this... One solution we discussed was adding a field to the Sample Details, that contains information of the pipeline that was used.

[arielsvn]

https://github.com/AlexsLemonade/refinebio/pull/299 Adds a pipelines field to each sample. This field will be an array of strings with the following possible values:

1. Submitter-processed
2. Affymetrix SCAN
3. Salmontools
4. Agilent SCAN TwoColor
5. Illumina SCAN
6. tximport
7. Salmon
8. MultiQC

@Miserlou had the idea of showing multiple values in the Processing Information field, for the samples that have more than one pipeline.

[arielsvn]

We talked about this issue, and for Keytar Kurt these are the possible values of pipeline for which we'll be able to show information.

• ['Submitter-processed']
• ['Affymetrix SCAN']
• ['Illumina SCAN']
• ['tximport', 'Salmon']

For the others, if they appear, we'll just show the text on the Processing Information column.

Also, https://github.com/AlexsLemonade/refinebio/issues/229 contains information about versions.

[arielsvn]

https://github.com/AlexsLemonade/refinebio-frontend/pull/50 Adds the processing information dialogs, but the package version information is static until the endpoints are exposed on the backend. AlexsLemonade/refinebio#229

[arielsvn]

https://github.com/AlexsLemonade/refinebio-frontend/pull/50 Added the last two processing information modals that we planned. Now we need to add version information. I opened #65 to keep track of that, so I'm closing this issue.