The paper presents a method called HHsearch for detecting distant homologous relationships between proteins using profile hidden Markov models (HMMs). The method is benchmarked against other tools like BLAST, PSI-BLAST, HMMER, and profile-profile comparison tools. HHsearch outperforms these tools in sensitivity, alignment quality, and speed. It uses profile HMMs instead of simple profiles, which improves sensitivity and alignment quality. HHsearch also incorporates predicted secondary structure information, which further enhances its performance. The method is efficient, with a query of 200 residues scanned against 3691 domains in 33 seconds on an AMD64 2GHz PC. HHsearch is available for download and includes up-to-date versions of SCOP and PFAM. The tool is used for homology detection, structure prediction, and protein evolution studies. The results show that HHsearch detects significantly more homologs than other methods, with higher alignment quality and faster performance. The method is also used in structure prediction servers and has been applied in blind structure prediction contests. The paper concludes that HHsearch represents a significant improvement over existing methods in terms of sensitivity and alignment quality.The paper presents a method called HHsearch for detecting distant homologous relationships between proteins using profile hidden Markov models (HMMs). The method is benchmarked against other tools like BLAST, PSI-BLAST, HMMER, and profile-profile comparison tools. HHsearch outperforms these tools in sensitivity, alignment quality, and speed. It uses profile HMMs instead of simple profiles, which improves sensitivity and alignment quality. HHsearch also incorporates predicted secondary structure information, which further enhances its performance. The method is efficient, with a query of 200 residues scanned against 3691 domains in 33 seconds on an AMD64 2GHz PC. HHsearch is available for download and includes up-to-date versions of SCOP and PFAM. The tool is used for homology detection, structure prediction, and protein evolution studies. The results show that HHsearch detects significantly more homologs than other methods, with higher alignment quality and faster performance. The method is also used in structure prediction servers and has been applied in blind structure prediction contests. The paper concludes that HHsearch represents a significant improvement over existing methods in terms of sensitivity and alignment quality.