The paper presents a method for predicting functional coupling between genes based on the conservation of gene clusters across multiple genomes. This approach uses conserved gene clusters to infer functional relationships, which is particularly useful for identifying the functions of uncharacterized genes in sequenced genomes. The method is based on the concept of "bidirectional best hits" (BBHs) and "pairs of close bidirectional best hits" (PCBBHs), which are defined as pairs of genes that are closely related and occur in the same gene cluster across multiple genomes. The study analyzed over 30 complete or nearly complete prokaryotic genomes and found that gene clusters are a common feature in bacterial chromosomes. These clusters often consist of genes that are functionally related, and their conservation across different species suggests that they are beneficial for the organisms that possess them. The method was tested on a variety of genomes, including both bacterial and archaeal species, and it was found that gene clusters play an important role in the Archaea as well. The study also introduced the concept of "pairs of close homologs" (PCHs), which are similar to PCBBHs but do not require the genes to be BBHs. This generalization allows for the detection of a broader range of functionally coupled genes. The results showed that the number of PCBBHs increases with the number of genomes analyzed, and that the coupling scores between genes can be used to infer functional relationships. The method was applied to reconstruct two common metabolic pathways: purine biosynthesis and glycolysis. The results demonstrated that the method can accurately predict functional couplings between genes, even for hypothetical proteins. The study also highlighted the importance of analyzing a large number of genomes to detect functional couplings, as the number of required genomes increases with the size of the functional subsystem. The paper concludes that the use of conserved gene clusters to infer functional coupling is a powerful tool for understanding the functions of genes in sequenced genomes. This approach has the potential to significantly contribute to the annotation of uncharacterized genes and the reconstruction of functional subsystems in prokaryotic genomes.The paper presents a method for predicting functional coupling between genes based on the conservation of gene clusters across multiple genomes. This approach uses conserved gene clusters to infer functional relationships, which is particularly useful for identifying the functions of uncharacterized genes in sequenced genomes. The method is based on the concept of "bidirectional best hits" (BBHs) and "pairs of close bidirectional best hits" (PCBBHs), which are defined as pairs of genes that are closely related and occur in the same gene cluster across multiple genomes. The study analyzed over 30 complete or nearly complete prokaryotic genomes and found that gene clusters are a common feature in bacterial chromosomes. These clusters often consist of genes that are functionally related, and their conservation across different species suggests that they are beneficial for the organisms that possess them. The method was tested on a variety of genomes, including both bacterial and archaeal species, and it was found that gene clusters play an important role in the Archaea as well. The study also introduced the concept of "pairs of close homologs" (PCHs), which are similar to PCBBHs but do not require the genes to be BBHs. This generalization allows for the detection of a broader range of functionally coupled genes. The results showed that the number of PCBBHs increases with the number of genomes analyzed, and that the coupling scores between genes can be used to infer functional relationships. The method was applied to reconstruct two common metabolic pathways: purine biosynthesis and glycolysis. The results demonstrated that the method can accurately predict functional couplings between genes, even for hypothetical proteins. The study also highlighted the importance of analyzing a large number of genomes to detect functional couplings, as the number of required genomes increases with the size of the functional subsystem. The paper concludes that the use of conserved gene clusters to infer functional coupling is a powerful tool for understanding the functions of genes in sequenced genomes. This approach has the potential to significantly contribute to the annotation of uncharacterized genes and the reconstruction of functional subsystems in prokaryotic genomes.