Example from this project

In the assays object in the ISAjson.
Under otherMaterials
The characteristics value are being filled with the Attribute Title and not the Attribute value for that SEEK Sample.
Example here

The values Title, submission date, status, and accession are actually the Titles for the Attributes (or the characteristicType annotationValue, example here). In this example, the characteristic Title has values input by user, while submission date, status, and accession have no values input.

Sometimes, each attribute has an identifier (often an IRI from an ontology).
The goal is just to distinguish where the attribute name has come from (By putting the IRI somewhere for the attribute name)

The ability to lock a template so that no one can create a new template based on that.
The lock checkbox can be placed in the template creation and used by the creator.

The protocol used in the processSequence to convert a Source to a Sample is not declared in the protocols object.

Example from this Project

The repository Templates are being listed in alphabetical order in the Choose a template menu when creating a Study.
They should be listed according to the "group_order": listed in their metadata.

In the left side section

• horizontal scrollable bar
• vertical scrollable bar
• show all items collapsed by default?
• "collapse all investigations" option somewhere
• "View more" button when there are too many Investigations

Similar to issue #34
Currently the exported ISAjson does not contain the SEEK ID of the Investigation.
That information can be added to the same comments object in investigation as the ISAjson export time is currently held in.
Example here.

A couple of comments can be added:
"name": "Investigation SEEK ID with "value": being the SEEK ID of the Investigation exported

In Single page view, on the Details tab in Study and in Assay, the field for Applied template (in yellow bellow) displays the name of the Template applied, but does not link to the Template.

It would be nicer if the name of the Template was also a link to that Template's view page.

In sample_type, the ontology name is the same of the attribute name, but it should be the name of the ontology instead, as specified in template.

SEEK sample ID could be shown in a column that precedes the Material column (xxx Name) in CSV export. Example below.

When concluding the creation of an asset in Single Page view, the user is currently redirected to the single page view at the Project level. It would be more convenient if the user was redirected to the level of the asset that was just created.

The @id values declared for each protocol in protocols does not match the ones being used in processSequence.

Example at protocols
Same at processSequence

Example from this Project

A Sample is being referenced as a other_material under derivesFrom in otherMaterials

Example from this project

Since @id are generally described as: "@id": { "type": "string", "format": "uri" }, it might be best to substitute spaces with _.
The substitution with _ would also help to maintain some readability for humans.

The @id in previousProcess and derivedProcess are incomplete in comparison to the originally declared @id.

Add the export time of the ISAjson file to the Investigation being exported.
Currently there is no indication inside of the .json exported ISAjson file of when the export happened. Since Investigations can change over time, having a timestamp inside of the file can help to solve potential confusion rising from two exports of the same Investigation having different contents.

The ISAjson export time can be added as a comment to the investigation following the investigation and comment schemas.

An example of a comment in an investigation can be found here
In the case of export time, we could have:
name: "ISAjson export time"
The value for value: being the export time down to the minutes, following ISO 8601 with a UTC offset indication.

In the following project
Input for Assay sequencing step (purple arrow) should be from library step, but it is from the Extraaction Assay (red arrow).

The study filename can be populated by the name given by the user to the Study in SEEK. Following the pattern s_StudyNameByUser.txt

Please add the ability to edit a Template that has no derived Sample Types.

• Necessary to fix typos and other mistakes only noticed after Template creation.
• Necessary for changing a not yet used Template after receiving feedback from a collaborator.
  • Example: “I made this Template, do you think I should add or remove any of the Attributes or does it look good as it is?”.

“Edit Template” can be under the Action button, together with “Delete Template”

When creating a derived Template (create Template based on existing Template), display a warning message when an Attribute in the parent Template that is marked as Required is modified.
Modified includes: complete deletion of that Attribute, change of Attribute Type, unchecking of the Required box, change in Attribute Name, etc.

Considering issue #23 , the value for an ISAjson Assay's @id cannot be the SEEK ID of a single SEEK Assay.
Example of an Assay @id in the current ISAjson export
Example from this Project

A solution might be to concatenate the SEEK IDs of all SEEK Assays that make up that ISAjson Assay.
For an ISAjson Assay that is made by SEEK Assays 13, 14, and 15, the @id would look like: "@id": "#assay/13_14_15",

From Project 17

For sample characteristic 3 the Ontology is about the attribute values (marked in purple below), but it is being listed as the definitions of the Attribute Title (PID, marked in yellow below).

The same mistake is made for the characteristicCategories of assays. Example here in which all of the listed Ontology sources are in reality for the values of the attributes and not for the titles.

    "materials": {
      "sources": [
        {
          "@id": "#source/71",
          "name": "plant 2",
          "characteristics": {
            "category": { "@id": "#characteristic_category/organism" },
            "value": {
              "annotationValue": "Pinus resinosa",
              "termSource": "efo",
              "termAccession": "http://purl.obolibrary.org/obo/OBI_0100026"
            },
            "unit": { "@id": null }
          }
        },

annotationValue": "Pinus resinosa" should have termAccession": "http://purl.obolibrary.org/obo/NCBITaxon_54921", and not "termAccession": "http://purl.obolibrary.org/obo/OBI_0100026" which correspond to "Organism".

When creating a Template, allow User to change the order of Attributes by dragging and dropping Attributes. Similar to what is possible when creating or editing a Sample Type.

The ontologySourceReferences gets populated upon ISA JSON file export, and it means the applied terms are not coming from the version necessarily.
Solution: Store ontologies information (name, file, version, description) when terms are getting populated

• The identifier in the study can be used to have accession numbers of data repositories.

• The submissionDate is also generally used to refer to the submission to a data repository

• The publicReleasDate is also generally used to refer to the public release in a data repository

We might want to leave these empty for now and not store SEEK study numbers there.

A otherMaterial here is referenced in samples with an non-existing sample prefix here.
It should not be in the samples object and should be called by its original otherMaterial @id

@rabuono I think we need to figure out what's best for SOP and then make specific issues about it.

        "protocols": [
            {
                "@id": "#protocol/seek id of the attribute with isaTag=protocol from template", //Probably best to have SEEK ID of the SOP assigned as Protocol for that Sample Type
                "name": "name of the attribute with isaTag=protocol from template", //Probably best is to use SOP title as discussed since it reflects User's name for protocol
                "protocolType": {
                    "annotationValue": "isa_protocol_type from template", //We have been adding these as the "Attribute name" of the attribute with isatag=protocol. For example "sample collection".
                    "termAccession": "iri of the attribute with isaTag=protocol from template",//This is about the title of the attribute, if defined by an ontology term. 
                    "termSource": "abbreviated ontology name (from the iri), as defined in ontologySourceReferences above"
                },
                "description": "description of the attribute with isaTag=protocol from template", //Originally this was supposed to be SOP description
                "uri": "iri again", //I am not sure SEEK has a IRI field for registering SOPs
                "version": "",
                "parameters": [
                    {
                        "@id": "#parameter/seek id of the attribute with isaTag=parameter_value from template", //Is associated with Sample Type, not Template.
                        "parameterName": {
                            "annotationValue": "attribute name of the attribute",
                            "termSource": "abbreviated ontology name (from the iri), as in ontologySourceReferences above",
                            "termAccession": "iri of the attribute with isaTag=parameter_value from template" //IRI of the term used as title for the attribute, if defined by an ontology term
                        }
                    }
                ],

"description": "description of the attribute with isaTag=protocol from template", //Originally this was supposed to be SOP description. Could we discuss if maybe we could pre-populate the protocol attribute with the SOP title? Or How do we use this attribute in dynamic table?

A unit @id value is being given in multiple places (example here) despite no units being declared in unitCategories.
Example from this Project

Note: At this point there is no way to add units to any of the attributes in the dynamic table. So there are not units to declare. The error might be on generating values for the @id of units when no units are rightfully not declared.

Assay n2 in a Study

• Assay design --> Samples: no samples shown
• Assay design --> Table: samples are shown
• No SOP associates

Assay --> Actions --> Delete assay: it gives error (This is not a secure form and http://0.0.0.0:3002/assays/15?return_to=%2Fsingle_pages%2F9)

Screenshots of the error:

Add the parent template field to the templates so that they can be listed base on the parent.

If a user inputs an invalid ontology term in DT, it has termAccession and termSource that come from the attribute ontology.
Expected: If the value is invalid, termAccession and termSource must be empty strings.
Solution: set termSource empty when termAccession is empty

In DT, I am not able to visualise the suggested terms (Input) for the 6th sample/row. DT doesn't scroll down anymore.

In an Investigation, if two attributes with isaTag other_material_characteristic have the same attribute title, they end up receiving the same @id in the ISAjson export. This causes the @id to not be unique.

An example from this project.
#characteristic_category/status appears three times: here, here and here.
These reflect an attribute of title status in the Source table and in two SEEK Assays.

When creating a Template based on existing Template, show all Attributes of parent Template as pre-populated items in the Attribute section.
It would be similar to a pre-populated creation of a new Template. Or similar to what one gets when editing an existing (and unused) Sample Type.

As there are no Streams in the current implementation, all the DataHub Assays belonging to a Study are to be combined into a single Assay in the ISAjson export in the assays object.
Each DataHub Assay corresponds to a step in the processSequence in that ISAjson Assay.

The Attributes tagged as source_characteristic seem to making their way into the studies characteristicCategories but are not all used in the characteristics object in source. All of the source_characteristics attributes used to describe a source should appear in the characteristics of that source.
Example from this Project

ISA Sources (111,112,113,114) are listed under assays --> materials --> samples instead of ISA samples (115,116).
Example here..

It should be:

"materials": {
              "samples": [
                { "@id": "#sample/115" },
                { "@id": "#sample/116" }
              ],

To avoid confusion when navigating and searching, but also conflict in an ISAjson export, it would be good if DataHub would enforce unique Study names linked to a same Investigation.
Otherwise multiple different studies with the same title and filename (in the ISAjson file) might be referred to in a same Investigation.

The is_title field will be used to define a sample_type title that is derived from a template.

From Project 17
'Source Characteristic 3` is not being exported to the ISAjson.

'Source Characteristic 2' and 'Source Characteristic 1' are exported.

For the SEEK creation date of a study, it would be better to have the date following ISO 8601 with a UTC offset indication. Similar to what is done for Investigation. Example here.

Similar to issue #30

Using a comments object inside each study in studies to record SEEK information about that study.
For now, add:
name: "SEEK ID"
value: The SEEK ID for that study (currently the URL)

name: "SEEK creation date"
value: The creation date recorded in SEEK for that study

Currently the exported ISAjson carries no information about which SEEK Project that Investigation belongs to.
That information can be added to the same comments object in investigation as the ISAjson export time is currently held.
Example here.

A couple of comments can be added:
"name": "SEEK Project name with "value": being the name of the Project to which the Investigation is associated with
"name": "SEEK Project ID with "value": being the SEEK ID for the Project to which the Investigation is associated with

the Comments part under the data_file is always []

In the ISAjson export the SEEK ID for the Investigation and the Study are given as 'comments' inside the appropriate object.

Currently for Investigation it is, for example:
"name": "SEEK Investigation ID", "value": 16

For Study:
"name": "SEEK ID", "value": "studies/16"

Might be best to have the studies closer to what is done for Investigation. On the same example as above, it would look like:
"name": "SEEK Study ID", "value": 16

When creating a Template based on an existing Template, the first item when in the Based on existing Template tab is the choice of the parent Template (e.g. a dropdown menu of existing Templates the User has access to).

By first choosing the parent Template, the User allows SEEK to fetch the correct information to pre-populate fields in the Template creation form (e.g. Template Description, and the list of Attributes with associated information).

Not all the SEEK's Assays attributes are listed under Assay characteristicCategories. These should be re-used under ISAjson assay --> materials --> otherMaterials -->characteristics --> category for each "otherMaterial".

Similar to #20

Rule number 28 of Ontology Annotations with a term and/or accession MUST provide a Term Source REF pointing to a declared Ontology Source Reference.
This conflicts with letting the user input free text for an ontology.