You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The way tools handle the ARC has gradually grown beyond the current ISA and ARC specifications. I composed a list of status-quo differences in the tools and the current state of the specification. These should be discussed and implemented into the ARC-specification:
Changes to ISA-Tab
Xlsx vs txt files
Investigation file keys:
Investigation PubMed ID changed to Investigation Publication PubMed ID
Study PubMed ID changed to Study Publication PubMed ID
Assay files now contain a metadata sheet with the assay section from the study section and an additional performers section
Study files now contain a metadata sheet with only the top-level study section
Sheetnames: isa_study, isa_assay and isa_investigation
Annotation table process separation
Process separation in ISA-Tab is marked using the Protocol REF column. Now, instead this separation is done by having different sheets
Annotation table sheet names correspond to isa-json process names
Annotation Table headers:
Space between category and name was introduced
e.g. Characteristics [Organism] vs Characteristics[Organism]
Ontological Annotation of headers:
E.g. Term Source REF (PATO:0000146) vs only Term Source REF
Added Component columns
Characteristic instead of Characteristics
Factor instead of Factor Value
Parameter isntead of Parameter Value
Entity headers now specify whether it’s Input or Output:
E.g. Input [Source Name] and Output [Sample Name]
Further Specifications
Paths
Based on arc root
Or relative to reference path:
Data file paths in assay files: assays/<assayName>/dataset
Data file paths in study files: studies/<studyName>/resources
Assay file name in metadata section: assays
Study file name in metadata section: studies
Parsing behaviour
Processes are parsed in a different way compared to ISA-API
In ISA-API, the input and output of the resulting process can be interpreted like in a chemical formula. All inputs map interchangeably to all outputs.
Currently in ARCtrl, the table logic is continued in the results isa_process.json. The n-th input maps to the n-th output in the lists
The text was updated successfully, but these errors were encountered:
The way tools handle the ARC has gradually grown beyond the current ISA and ARC specifications. I composed a list of status-quo differences in the tools and the current state of the specification. These should be discussed and implemented into the ARC-specification:
Changes to ISA-Tab
Xlsx vs txt files
Investigation file keys:
Investigation PubMed ID
changed toInvestigation Publication PubMed ID
Study PubMed ID
changed toStudy Publication PubMed ID
Metadata sheets
isa_study
,isa_assay
andisa_investigation
Annotation table process separation
Protocol REF
column. Now, instead this separation is done by havingdifferent sheets
Characteristics [Organism]
vsCharacteristics[Organism]
Term Source REF (PATO:0000146)
vs onlyTerm Source REF
Component
columnsCharacteristic
instead ofCharacteristics
Factor
instead ofFactor Value
Parameter
isntead ofParameter Value
Input [Source Name]
andOutput [Sample Name]
Further Specifications
Paths
assays/<assayName>/dataset
studies/<studyName>/resources
assays
studies
Parsing behaviour
Processes are parsed in a different way compared to ISA-API
The text was updated successfully, but these errors were encountered: