Phyre2 is a web-based tool for predicting and analyzing protein structure, function, and mutations. It provides a user-friendly interface to advanced bioinformatics methods, replacing the previous Phyre server. Phyre2 uses remote homology detection to build 3D models, predict ligand binding sites, and analyze the effect of amino acid variants. Users can submit a protein sequence and receive results detailing the secondary and tertiary structure, domain composition, and model quality. Additional tools allow users to search for protein structures in a genome, submit large numbers of sequences, and automatically scan for difficult-to-model proteins. The server is available at http://www.sbg.bio.ic.ac.uk/phyre2 and typically returns results within 30 minutes to 2 hours. Phyre2 is one of the most widely used protein structure prediction servers, serving approximately 40,000 unique users per year and processing 700–1,000 user-submitted proteins daily. It has been applied to the annotation of a wide range of genomes. Phyre2 is compared with other structure prediction servers, such as i-TASSER, Swiss-Model, HHpred, and Robetta. While these tools are generally similar in accuracy for most modeling tasks, Phyre2 is noted for its ease of use and user-friendly interface. In international blind trials (CASP), Phyre2 ranked among the top performers, with improvements in model quality compared to other methods. Phyre2 has limitations, including the inability to model proteins without homology to known structures and the inability to accurately predict the structural effects of point mutations. It also lacks the ability to model multimers and may produce incorrect domain orientations when combining separate domains. Phyre2 is a combination of various software components and runs on a Linux farm with 300 CPU cores. It offers advanced facilities such as Backphyre for searching structures across genomes, batch processing, one-to-one threading, and Phyrealarm for automated scanning of difficult-to-model proteins. Phyre Investigator is a tool for in-depth analysis of model quality, function, and mutation effects. Phyre2 allows users to submit a protein sequence and receive results including the top-scoring model, confidence in the model, and information about the structural template. The results are presented in a user-friendly format, with options for detailed analysis, including model quality assessment, alignment confidence, clash detection, and mutational analysis. Phyre2 is free to use, and users can access additional tools such as the Phyre2 job manager to track and renew jobs. The server is continuously updated with new structures and sequence databases, improving the detection of homology over time.Phyre2 is a web-based tool for predicting and analyzing protein structure, function, and mutations. It provides a user-friendly interface to advanced bioinformatics methods, replacing the previous Phyre server. Phyre2 uses remote homology detection to build 3D models, predict ligand binding sites, and analyze the effect of amino acid variants. Users can submit a protein sequence and receive results detailing the secondary and tertiary structure, domain composition, and model quality. Additional tools allow users to search for protein structures in a genome, submit large numbers of sequences, and automatically scan for difficult-to-model proteins. The server is available at http://www.sbg.bio.ic.ac.uk/phyre2 and typically returns results within 30 minutes to 2 hours. Phyre2 is one of the most widely used protein structure prediction servers, serving approximately 40,000 unique users per year and processing 700–1,000 user-submitted proteins daily. It has been applied to the annotation of a wide range of genomes. Phyre2 is compared with other structure prediction servers, such as i-TASSER, Swiss-Model, HHpred, and Robetta. While these tools are generally similar in accuracy for most modeling tasks, Phyre2 is noted for its ease of use and user-friendly interface. In international blind trials (CASP), Phyre2 ranked among the top performers, with improvements in model quality compared to other methods. Phyre2 has limitations, including the inability to model proteins without homology to known structures and the inability to accurately predict the structural effects of point mutations. It also lacks the ability to model multimers and may produce incorrect domain orientations when combining separate domains. Phyre2 is a combination of various software components and runs on a Linux farm with 300 CPU cores. It offers advanced facilities such as Backphyre for searching structures across genomes, batch processing, one-to-one threading, and Phyrealarm for automated scanning of difficult-to-model proteins. Phyre Investigator is a tool for in-depth analysis of model quality, function, and mutation effects. Phyre2 allows users to submit a protein sequence and receive results including the top-scoring model, confidence in the model, and information about the structural template. The results are presented in a user-friendly format, with options for detailed analysis, including model quality assessment, alignment confidence, clash detection, and mutational analysis. Phyre2 is free to use, and users can access additional tools such as the Phyre2 job manager to track and renew jobs. The server is continuously updated with new structures and sequence databases, improving the detection of homology over time.