The Author Curation to Knowledgebase project (ACKnowledge, previously known as “Author First Pass”) aims at facilitating author curation to biological knowledgebases.
Biological knowledgebases serve as a critical resource for researchers and accelerate scientific discoveries by providing manually curated, machine-readable data collections. At present, the aggregation and manual curation of biological data is a labor-intensive process that relies almost entirely on professional biocurators. Biocuration, however, is greatly aided when authors directly participate in the process.
ACKnowledge employs a semi-automated curation pipeline that identifies biological entities and data types (for example, genes and anatomic expression) from newly published research articles and then automatically emails authors to review the extracted information through a dedicated web interface. During the curation process, authors may modify extracted data, an important quality control measure. Participating authors are publicly acknowledged for their contribution in the knowledgebase.
The ACKnowledge system, currently implemented by WormBase for the C. elegans literature and soon to be implemented for other species, couples statistical methods and state-of-the-art text mining algorithms to identify entities and data types.
Additional tools in the ACKnowledge system include software to send automated reminder emails to authors who have not completed their submissions, a notification system to inform knowledgebase curators about new author submissions, and a dashboard for curators to monitor submission status.
The ACKnowledge project is supported by grant RO1 OD023041 from the National Library of Medicine at the US National Institutes of Health.
If you have questions about the ACKnowledge project, please contact help.acknowledge@wormbase.org.Release Notes
Version 4.0
Release date
10-27-2022
Introduction
Release 4.0 marks an exciting transition for our project. With support from the National Library of Medicine, the Author First Pass (AFP) project is now officially known as ACKnowledge (Author Curation to Knowledgebase). This change reflects our commitment to expanding artificial intelligence (AI)-enhanced community curation to capture more detailed experimental findings and to introduce our curation pipeline to additional model organism communities and their respective knowledgebases. While busy transitioning our project and enhancing our AI methods, we also implemented further improvements to the current pipeline based on continuing valuable feedback from authors. These improvements, listed below, will help authors add more information about their papers and readily find previously published papers processed by our pipeline for curation using our Author Curation Portal.
Main Updates
- Renamed project to ACKnowledge (Author Curation to Knowledgebase)
- Updated user facing communications and forms
- Updated internal documentation
- Created an ACKnowledge logo
- Created an ‘About’ widget on the author curation form
- Provided project description
- Added NLM funding information
- Listed published references
- Created a project landing page: https://wormbase.github.io/ACKnowledge
- Fully transitioned from Support Vector Machines (SVM) to Neural Networks (NN) for datatype flagging
User Updates
- Enabled access to author’s additional publications via the author curation portal (https://acp.acknowledge.textpressolab.com) from the author submission form and from all user-facing communications (emails)
- Added free text box for chemical- and environmental-induced phenotypes
- Added a free text box for adding new species not in the current auto-complete list
- Improved the comments section
Bug Fixes
- Form now saves incremental updates to genes and species lists to database without requiring a full resubmission
- Proof papers are now excluded by the processing pipeline
- Improved authors’ email extraction
- Additional minor bug fixes
Curator Dashboard Updates
- Improved statistics page to keep track of response rate on a yearly and monthly basis
- Standardized representation of empty values in curator dashboard
Version 3.0
Release Date
09-16-2021
Introduction
Based on valuable user feedback, we continue to improve the ACKnowledge experience. In addition to providing more streamlined ways to enter entities, we’ve added further instructions for adding new entities, and a link to look up specific entities in WormBase. We now flag papers for enzymatic activity and include a free-text field to describe other gene functions. We updated the FAQs and provide a link to the ACKnowledge webinar presented in February 2021. Lastly, to thank authors for their valuable contributions, we now acknowledge authors on WormBase person and paper pages.
We have implemented updates and improvements to our curator dashboard. The dashboard is a tool available to WormBase curators to keep track of submissions and perform quality control of submitted data.
Main Updates
User Updates
- Acknowledge author contributions on WormBase Paper and Person pages
- Email all authors of a publication whenever possible
- Added linkout to WB pages for entities in select components
- Simplified add entities lists
- Improved instructions for Allele, Strain, and Transgene sections
- Added WormBase strain IDs
- Expanded exclusion lists (e.g. M9 buffer not recognized as strain)
- Included enzymatic activity neural network
- Added other gene function checkbox
- Created link to 'WormBase ACKnowledge Webinar' video
- Updated the FAQ section
Curator Dashboard Updates
- Created summary page with sortable lists of entities removed and added by authors
- Generated a pre-populated spreadsheet of all flagged data for each processed paper to allow authors an alternative data submission pipeline
- Updated dashboard thresholds with new TFIDF values
- Sorted contributors by number of submissions
- Added download button for lists of papers
Version 2.0
Release Date
09-14-2020
Introduction
We improved the ACKnowledge system based on feedback received from authors and statistics collected during one year of ACKnowledge v1.0 author submissions. We improved the definition of data types in the form and we refined our text mining algorithms to obtain better results.
Main Updates
- Implemented TFIDF in place of simple thresholds to improve precision of gene and allele recognition
- Improved datatype descriptions in "?" mouse-overs
- Added FAQs and Release Notes
Version 1.0
Release date
06-03-2019
Introduction
We are proud to introduce our new ACKnowledge system, which leverages text mining to help authors curate their papers and send data to WormBase! The system replaces the previous form which required authors to submit data manually through free-text forms.
Main Updates
- Automated extraction of biological entities from articles based on text mining
- Extracted entities presented to author for validation through a newly designed web form
- Additional data types can be manually entered by authors
- Dashboard for curators to monitor submissions
- Monthly notifications to curators with newly submitted data
- Periodic email reminders to authors
- More details in our paper published on Database
References
[1] Arnaboldi V, Raciti D, Van Auken K, Chan JN, Muller H-M, Sternberg PW (2020) Text mining meets community curation: a newly designed curation platform to improve author experience and participation at WormBase. Database, doi:10.1093/database/baaa006