The article introduces PANTHER, a comprehensive database for classifying protein sequences. PANTHER consists of two main components: the PANTHER library (PANTHER/LIB) and the PANTHER index (PANTHER/X). PANTHER/LIB is a collection of protein families, each represented by a multiple sequence alignment, a Hidden Markov Model (HMM), and a family tree. Functional divergence within families is represented by dividing the tree into subtrees based on shared functions and by subtree HMMs. PANTHER/X is an abbreviated ontology for summarizing and navigating molecular functions and biological processes associated with the families and subfamilies. The authors apply PANTHER to three areas: characterizing the size and sequence diversity of protein families and subfamilies, visualizing the relative number of human and mouse genes with specific molecular functions or biological processes, and ranking missense single nucleotide polymorphisms (SNPs) according to their likelihood of affecting protein function. The results show that PANTHER can effectively classify proteins, provide insights into functional divergences, and predict the impact of SNPs on protein function.The article introduces PANTHER, a comprehensive database for classifying protein sequences. PANTHER consists of two main components: the PANTHER library (PANTHER/LIB) and the PANTHER index (PANTHER/X). PANTHER/LIB is a collection of protein families, each represented by a multiple sequence alignment, a Hidden Markov Model (HMM), and a family tree. Functional divergence within families is represented by dividing the tree into subtrees based on shared functions and by subtree HMMs. PANTHER/X is an abbreviated ontology for summarizing and navigating molecular functions and biological processes associated with the families and subfamilies. The authors apply PANTHER to three areas: characterizing the size and sequence diversity of protein families and subfamilies, visualizing the relative number of human and mouse genes with specific molecular functions or biological processes, and ranking missense single nucleotide polymorphisms (SNPs) according to their likelihood of affecting protein function. The results show that PANTHER can effectively classify proteins, provide insights into functional divergences, and predict the impact of SNPs on protein function.