PANTHER: Making genome‐scale phylogenetics accessible to all

• Published in: Protein science Vol. 31; no. 1; pp. 8 - 22
• Main Authors: Thomas, Paul D., Ebert, Dustin, Muruganujan, Anushya, Mushayahama, Tremayne, Albou, Laurent‐Philippe, Mi, Huaiyu
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.01.2022; Wiley Subscription Services, Inc
• Subjects: Amino acid sequence; Computer applications; Databases, Protein; Evolution; Evolution, Molecular; gene ontology; Genetic diversity; Genetic variance; genome analysis; Genomes; hidden Markov model; Homology; Knowledge bases (artificial intelligence); molecular evolution; Molecular Sequence Annotation; Nucleotide sequence; omics data analysis; phylogenetic tree; Phylogenetics; Phylogeny; protein function annotation; protein function evolution; Proteins; Proteins - chemistry; Proteins - genetics; Quality control; Review; Sequence analysis; Sequence Analysis, Protein; Software; Software development tools; Systematic errors; omics data analysis; gene ontology; phylogenetic tree; protein function annotation; protein function evolution; hidden Markov model; genome analysis; molecular evolution
• ISSN: 0961-8368; 1469-896X; 1469-896X
• DOI: 10.1002/pro.4218

Phylogenetics is a powerful tool for analyzing protein sequences, by inferring their evolutionary relationships to other proteins. However, phylogenetics analyses can be challenging: they are computationally expensive and must be performed carefully in order to avoid systematic errors and artifacts. Protein Analysis THrough Evolutionary Relationships (PANTHER; http://pantherdb.org) is a publicly available, user‐focused knowledgebase that stores the results of an extensive phylogenetic reconstruction pipeline that includes computational and manual processes and quality control steps. First, fully reconciled phylogenetic trees (including ancestral protein sequences) are reconstructed for a set of “reference” protein sequences obtained from fully sequenced genomes of organisms across the tree of life. Second, the resulting phylogenetic trees are manually reviewed and annotated with function evolution events: inferred gains and losses of protein function along branches of the phylogenetic tree. Here, we describe in detail the current contents of PANTHER, how those contents are generated, and how they can be used in a variety of applications. The PANTHER knowledgebase can be downloaded or accessed via an extensive API. In addition, PANTHER provides software tools to facilitate the application of the knowledgebase to common protein sequence analysis tasks: exploring an annotated genome by gene function; performing “enrichment analysis” of lists of genes; annotating a single sequence or large batch of sequences by homology; and assessing the likelihood that a genetic variant at a particular site in a protein will have deleterious effects.