TopFIND (1–3) is a protein centric database for the annotation of protein termini currently in its third version. Non canonical protein termini can be the result of multiple different biological processes, including pre translational processes such as alternative splicing and alternative translation initiation or post translational protein processing by proteases that cleave proteases as part of protein maturation or as a regulatory modification. Accordingly, protein termini evidence in TopFIND is inferred from other databases such as ENSEMBL (4) transcripts, TISdb (5) for alternative translation initiation, MEROPS (6) for protein cleavage by proteases, and UniProt (7) for canonical and protein isoform start sites. Additionally, termini are annotated from user submitted lists of termini and inferred from user submitted lists of cleavage sites.