SUMMARY: Large numbers of rare and unique titin missense variants have been discovered in both healthy and disease cohorts, thus the correct classification of variants as pathogenic or non-pathogenic has become imperative. Due to titin's large size (363 coding exons), current web applications are unable to map titin variants to domain structures. Here, we present a web application, TITINdb, which integrates titin structure, variant, sequence and isoform information, along with pre-computed predictions of the impact of non-synonymous single nucleotide variants, to facilitate the correct classification of titin variants. AVAILABILITY AND IMPLEMENTATION: TITINdb can be freely accessed at http://fraternalilab.kcl.ac.uk/TITINdb. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.