How can bioinformatics tools help to identify particular types of proteins? In general, the answer depends on the type of protein. Alignment methods can identify similarities between two proteins. However, although database search tools are optimized to finding the best possible superposition between two proteins, they fail in answering questions such as: Does the query protein Q perform the same function as the protein in the database H for which we have some experimental data about function? In fact, alignment methods typically provide some statistical score evaluating the probability that the similarity between Q and H happened by chance (1 ,2 ). The precise function relating such a statistical score for sequence similarity to the actual biological similarity of two proteins, that is, similarity in terms of their three-dimensional (3D) structure and/or function depends on the problem. For example, if the Position-Specific Iterated Basic Local Alignment Search Tool (PSI-BLAST [2 ]) expectation value for the similarity between Q and H is below 10−5 , then this typically implies that H and Q have similar local 3D structure (3 ). However, less than 70% of all pairs of enzymes that have this level of sequence similarity have exactly the same enzymatic activity (4 ), and over 90% of all pairs with so similar sequences are observed in the same subcellular compartment (5 ). Establishing these estimates typically requires solving three different tasks: (a) defining biological similarity (3D, enzyme activity, subcellular localization); (b) building unbiased data sets of experimentally reliable information; and (c) establishing thresholds that relate sequence to biological similarity.