The sum of estimated annual payments per person for inpatient hospital admissions, annual payments per person for outpatient services, annual payments per person for medicines, and annual payments per person for glucose-testing supplies. To avoid doublecounting, because self-reported payments for OPVs and admissions included payments for medicines, the study team subtracted from the grand total payments for medicines and strips that were purchased from hospitals during visits and admissions. To calculate the amount to subtract, the team first estimated the mean “as-used” supply of medicines, including aspirin and other over-the-counter products, when purchased from hospital pharmacies. Based on patient self-report, we then calculated the proportion of hospital clinic OPVs during which medicines were purchased. Because the interview schedule did not ask subjects which medicines they purchased at OPVs, we assumed that one refill of every current medicine was purchased at every OPV. To make use of these data, comparative analysis has often been used to induce meaningful hypotheses through discovery of conserved sequences with regulatory functions and novel genes. Each protein contains domains that have unique functions and can evolve independently of the rest of the protein chain. A domain is generally considered as a compact and semi-independent unit that can fold into a PB 203580 cost stable, three-dimensional structure. Such evolutionary relationships between closely related species can be revealed by comparative analysis of their domains. The prediction of protein domains has long been considered one of the most fundamental steps in deciphering the evolution and functions of proteins as well as species. Domain detection is often closely related to the determination of discrete structural folding units. Various domain prediction methods have been reported in the literature. The existing methods can be classified into two main categories, namely template-based methods and de novo methods. The template-based methods identify the similarities between a target sequence and the template sequences in a protein structure database such as Protein Data Bank. However, the accuracy of the template-based methods is highly dependent on the quality of the template structures. Therefore, such methods should not be assumed to work well for proteins containing novel domains, especially when they are from less characterized species. On the other hand, the ab-initio methods can predict protein domains by taking advantage of various sequence-based features, including sequence profiles, secondary structure predictions, and correlated mutations. Those methods use computational tools, such as neural networks, support vector machines, and hidden Markov models. However, the accuracy of ab-initio domain prediction methods on multidomain proteins is still very low. All these methods have either a well-defined structural database or structure-related features as their foundations. However, structural information is available for only a very tiny fraction of the entire set of proteins.