Application Notes in JCIM: A Home for Data, Software, and Web-Based Services in Cheminformatics and Molecular Modeling
Ariane Nunes Alves, Rommie E. Amaro, Yee Siew Choong, Zoe Cournia, Xiao He, Olexandr Isayev, Johannes Kirchmair, John Kitchin, Kevin J. Naidoo, Duc Nguyen, Giulia Palermo, Thereza A. Soares, Guo-Wei Wei, Tong Zhu and Kenneth M. Merz Jr., “Application Notes in JCIM: A Home for Data, Software, and Web-Based Services in Cheminformatics and Molecular Modeling”, Journal of Chemical Information and Modeling 66(17), 10447-10449 (2026). https://doi.org/10.1021/acs.jcim.6c02729
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Question
Curated databases, benchmark sets, software, web services and integrated platforms increasingly shape how research in the chemical, life and materials sciences is done and validated. This editorial by the JCIM editorial team, of which Kitchin is a member, asks what the journal’s Application Note article type should cover and what standards such reports should meet (Nunes Alves et al. 2026).
Approach
The editorial is a short policy statement, not a research paper. It defines the Application Note, describes five categories of encouraged contributions with examples previously published in JCIM, and sets out formal requirements and review criteria, including new submission standards aimed at the longevity of published tools.
Key findings
An Application Note is a peer-reviewed report of a new or substantially extended digital resource, and its defining feature is that the resource must be usable by the community rather than a conceptual framework or internal workflow. Five categories are described. Databases should document scope, update strategy, standards and identifiers, curation and quality control, access mechanisms, and how they relate to existing resources (examples include ToxFCDB, NavDB and PharmGEO). Benchmark data sets should document provenance, inclusion criteria, curation and preprocessing, known limitations and biases, and recommended evaluation protocols with baseline results (examples include the Amber free energy validation set, UNIQUE, AbSet and ThermoPred). Software packages should state the problem addressed and the advance over existing tools, give example workflows, installation details, dependencies and supported platforms, and ideally be platform independent and available for academic and commercial use (AmberTools, Chemprop, REINVENT 2.0, SuAVE-Scatter). Web services should be publicly accessible at publication with clear input and output formats and an explicit plan for long-term availability (SwissParam, MolGpka, CHARMM-GUI Ligand Docker, PPB3). Integrated platforms that couple generative models, property prediction, docking and visualization must document each component and the architecture and address hosting sustainability (Chemistry42, PandaOmics, AI-MedCraft).
Formally, Application Notes are limited to 4000 words excluding references, with justified exceptions explained in the cover letter, must include at least one figure, and should ideally name the resource in the title. To guard against short-lived, AI-generated code, JCIM is tightening its standards: standalone packages must show an open repository with at least six months of active development before submission plus community uptake metrics such as downloads, and web services need a sustainability statement on guaranteed hosting and an open-source repository that would let others redeploy the service. FAIR data principles and clear licensing are strongly encouraged, the resource must be accessible to reviewers anonymously, and review weighs functionality, reliability, accessibility, documentation and likely community impact alongside scientific merit.
Significance
The editorial positions JCIM Application Notes as a primary venue for digital scholarship in cheminformatics and molecular modeling and makes sustainability and demonstrated maintenance explicit acceptance criteria. It builds on earlier JCIM and JCTC editorials on method and data sharing and reproducibility, including one Kitchin coauthored (amaro2026), and it complements his long-standing advocacy for sharing data and code with publications.