preDQ is a software tool for prediction of peptides binding to HLA‐DQ2 and HLA‐DQ8 proteins, associated with susceptibility to celiac disease. preDQ was designed to evaluate the risk of novel proteins to cause celiac disease following EFSA’s guidance on assessment of genetically modified plants. The predictive capability of preDQ relies on the utilization of five robust models, each producing five corresponding risk assessment outputs. These models have been developed using datasets containing both known binding and non‐binding peptides to HLA‐DQ2.5 and HLA‐DQ8.1. The models were derived by ligand‐based and structure‐based computational methods and validated by internal cross‐validation and external datasets. Only the highest‐performing models have been implemented in preDQ. Overall, preDQ represents a comprehensive, user‐friendly, and reliable tool for evaluating the binding affinity of peptides to HLA‐DQ2.5 and HLA‐DQ8.1 and the potential of the tested proteins to trigger celiac disease. Here, we propose a project aimed at updating, upgrading and maintaining of preDQ. As the authors of the software tool and the models that support it, we offer to update the datasets used to derive and validate the models, re‐evaluate the existing models using the new data and derive new models, as well as maintain the tool, provide consistent feedback to users and stakeholders, publish the derived models in peer‐reviewed journals and closely collaborate with EFSA experts.
