A large variety of molecular interactions occurs between biomolecular components in cells. When one or a cascade of molecular interactions results in a regulatory effect, by one component onto a downstream component, a so-called ‘causal interaction’ takes place. Causal interactions constitute the building blocks in our understanding of larger regulatory networks in cells. These causal interactions and the biological processes they enable (e.g., gene regulation) need to be described with a careful appreciation of molecular interactions that occur between entities. A proper description of this information enables archiving, sharing, and reuse by humans and for computational science. Various representations of causal relationships between biological components are currently used in a variety of resources. Here, we propose a checklist that accommodates current representations, and call it the Minimum Information about a Molecular Interaction CAusal STatement (MI2CAST). This checklist defines both the required core information, as well as a comprehensive set of other contextual details valuable to the end user and relevant for reusing and reproducing causal molecular interaction information. The MI2CAST checklist can be used as reporting guidelines when annotating and curating causal statements, while assuring uniformity and interoperability of the data across resources.
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Received:
29 April 2020
Commenter:
Charles Tapley Hoyt
(Click to see Publons profile: )
The commenter has declared there is no conflict of interests.
Comment:
After acknowledging that PSI-MITAB2.8, BEL, GO-CAM all cover the full range of metadata recommended by MI2CAST, it would be good to give examples how how this is done in each format. Otherwise, this still seems quite abstract.
Later, the article suggests that PSI-MITAB2.8 is compliant. I'm sure to many readers, including myself, the nuances of all of the PSI-*** formats remain elusive. It would also be beneficial to give an example. For people who work with data, giving a structured definition of the schema for this format would also be beneficial. Minimally, a link to a page with more information should be provided.
Throughout, the article switches between referencing entities with CURIEs and PURL links. It might be stylistically beneficial to stick to one.
Received:
30 April 2020
Commenter:
Vasundra Touré
The commenter has declared there is no conflict of interests.
Comment:
Thank you for your feedback Charles!
We will provide examples in the different formats in the GitHub repository. A table would also be useful to highlight which terms are supported in which 'field' of that format. We will add this as well.
The three formats you've cited do not all support the full range of metadata recommended by MI2CAST. The sentence is stated the other way around: MI2CAST contains recommendations of terms that are possible to be annotated in at least one of these formats. We will reformulate this to make it clearer. The hope is the different formats will, in the future, enable the annotation and storage of all information content recommended in MI2CAST.
Regarding PSI-MITAB2.8, we will add a link to their GitBook that explicit the different columns' content: psicquic.github.io/MITAB28Format.html. There you can also find the different variants (2.7, 2.6, etc.). The PSI-MITAB2.8 manuscript (doi.org/10.1093/bioinformatics/btz132) is also a good reading to learn about the specific changes done in PSI-MITAB2.8 to support causality. In MI2CAST, we provide some examples of PSI-MITAB2.8 files: github.com/MI2CAST/MI2CAST/tree/master/examples.
Regarding CURIES and PURL links: Good point! Thanks, we'll fix it.
Vasundra Touré
*
,
Commenter: Charles Tapley Hoyt (Click to see Publons profile:
The commenter has declared there is no conflict of interests.
Later, the article suggests that PSI-MITAB2.8 is compliant. I'm sure to many readers, including myself, the nuances of all of the PSI-*** formats remain elusive. It would also be beneficial to give an example. For people who work with data, giving a structured definition of the schema for this format would also be beneficial. Minimally, a link to a page with more information should be provided.
Throughout, the article switches between referencing entities with CURIEs and PURL links. It might be stylistically beneficial to stick to one.
Commenter: Vasundra Touré
The commenter has declared there is no conflict of interests.
We will provide examples in the different formats in the GitHub repository. A table would also be useful to highlight which terms are supported in which 'field' of that format. We will add this as well.
The three formats you've cited do not all support the full range of metadata recommended by MI2CAST. The sentence is stated the other way around: MI2CAST contains recommendations of terms that are possible to be annotated in at least one of these formats. We will reformulate this to make it clearer. The hope is the different formats will, in the future, enable the annotation and storage of all information content recommended in MI2CAST.
Regarding PSI-MITAB2.8, we will add a link to their GitBook that explicit the different columns' content: psicquic.github.io/MITAB28Format.html. There you can also find the different variants (2.7, 2.6, etc.). The PSI-MITAB2.8 manuscript (doi.org/10.1093/bioinformatics/btz132) is also a good reading to learn about the specific changes done in PSI-MITAB2.8 to support causality. In MI2CAST, we provide some examples of PSI-MITAB2.8 files: github.com/MI2CAST/MI2CAST/tree/master/examples.
Regarding CURIES and PURL links: Good point! Thanks, we'll fix it.