Meta-MEME: motif-based hidden Markov models of protein families

Sirolimus is a new immunosuppressive drug that has been evaluated in animal experiments. The current study was conducted on humans with reformulated sirolimus in doses from 3 mg/m2 to 15 mg/m2. Sixteen renal transplant recipients were included in this phase I study to determine the safety, tolerance, and preliminary pharmacokinetics of increasing single doses of orally administered sirolimus. All 16 patients had stable renal graft function after a renal transplant at least 6 months before the study. Basal immunosuppression consisted of cyclosporine and prednisolone (n = 10) or cyclosporine, azathioprine, and prednisolone (n = 6). Four groups (I, 3 mg/m2; II, 5 mg/m2; III, 10 mg/m2; IV, 15 mg/m2) of four patients were assigned randomly to receive sirolimus (n = 3) or placebo (n = 1). Among the 12 patients who received sirolimus, five had mild transient study events such as headache, nausea, mild dizziness, hypoglycemia, epistaxis, and decrease in platelets. No serious adverse events occurred and no nephrotoxic effects could be related to the single dose administration of sirolimus. The only study event that was judged as probably related to sirolimus was the single case of thrombocytopenia. The other events were evaluated as possibly related. Thrombocytopenia occurred at the highest dose level (15 mg/m2 sirolimus). In two of the patients in the placebo group, slight elevations of liver enzymes and serum amylase were seen. Blood and plasma sirolimus concentrations were analyzed by an electrospray-high performance liquid/mass spectrophotometric (ESP-HPLC/MS) method Sirolimus showed an extensive red blood cell distribution with a mean blood/ plasma ratio of 49.1. The elimination half-life ranged from 43.8 to 86.5 hours (mean 56.9 hours). The Cmax and the area under the concentration versus time curves (AUC) correlated reasonably with doses from 3 to 15 mg/m2. The oral dose clearance ranged from 42 to 339 ml/h.kg. No clinically significant differences were seen in the trough concentrations of cyclosporine or the AUCs before and after the administration of sirolimus. Administration of single oral doses of sirolimus from 3 to 15 mg/m2 was safe and well tolerated in stable renal transplant recipients. Thrombocytopenia may be the dose-limiting toxicity. Additional phase II and phase III clinical trials will define the immunosuppressive efficacy of sirolimus.     

Weighting hidden Markov models for maximum discrimination

MOTIVATION: Hidden Markov models can efficiently and automatically build statistical representations of related sequences. Unfortunately, training sets are frequently biased toward one subgroup of sequences, leading to an insufficiently general model. This work evaluates sequence weighting methods based on the maximum-discrimination idea. RESULTS: One good method scales sequence weights by an exponential that ranges between 0.1 for the best scoring sequence and 1.0 for the worst. Experiments with a curated data set show that while training with one or two sequences performed worse than single-sequence Probabilistic Smith-Waterman, training with five or ten sequences reduced errors by 20% and 51%, respectively. This new version of the SAM HMM suite outperforms HMMer (17% reduction over PSW for 10 training sequences), Meta-MEME (28% reduction), and unweighted SAM (31% reduction). AVAILABILITY: A WWW server, as well as information on obtaining the Sequence Alignment and Modeling (SAM) software suite and additional data from this work, can be found at http://www.cse.ucse. edu/research/compbio/sam.html     

Profile hidden Markov models

The recent literature on profile hidden Markov model (profile HMM) methods and software is reviewed. Profile HMMs turn a multiple sequence alignment into a position-specific scoring system suitable for searching databases for remotely homologous sequences. Profile HMM analyses complement standard pairwise comparison methods for large-scale sequence analysis. Several software implementations and two large libraries of profile HMMs of common protein domains are available. HMM methods performed comparably to threading methods in the CASP2 structure prediction exercise.     

Heart signal recognition by Hidden Markov Models: the ECG case

From December 1989 to May 1990, 315 faecal samples from children under 5 years old with diarrhoea (215) and without diarrhoea (100) seen at paediatric clinics were investigated for bacterial, viral and parasitic enteropathogens. Standard and recently described methods were used for the investigations, which revealed that 74.9% of children with diarrhoea were infected with enteropathogens compared with 28% of controls. In the diarrhoeal group, 59.1% had a bacterial, 26.5% a viral and 2.3% a parasitic aetiology. Rotavirus was the pathogen most frequently detected, accounting for 22.3% of positive findings in the group with diarrhoea versus 9% in the control group. Other important agents were: enterotoxigenic Escherichia coli (ETEC) (14.4 versus 6%), enteropathogenic E. coli (EPEC) (10.7 versus 5%), enteroadherent E. coli (EAEC) (9.3 versus 4%), enterohaemorrhagic E. coli (EHEC) (5.1 versus 3%) and Salmonella spp. (3.3 versus 1%). The following enteropathogens were detected exclusively in the diarrhoeal stools: Shigella spp. (5.1%), Yersinia enterocolitica (0.9%), Aeromonas hydrophila (1.4%), Entamoeba histolytica (0.5%), Giardia lamblia (0.5%), Trichomonas hominis (0.5) and Trichuris trichiura (0.9%). The detection rates of rotavirus, EPEC and EAEC were much greater in the diarrhoeal than in the control patients. No Vibrio cholerae, enteroinvasive E. coli (EIEC), Plesiomonas spp. or Cryptosporidium spp. were detected in this study. Our data suggest that both the traditional and newly recognised diarrhoeal agents are important causes of diarrhoea in the children under 5 years old in Lagos, Nigeria.     

Use of latent variable models for detecting discrimination in salaries.

A major criticism of the direct regression approach to detecting employment discrimination has been that when qualifications are not perfectly measured, one could be misled by the well-known bias associated with using fallible covariates with nonequivalent groups—sometimes called the "errors in variables" problem. With regard to salary discrimination, a number of solutions to this "problem" have been proposed. I argue that this bias should be treated as irrelevant and that the proposed solutions are misleading. In order to sustain the argument against direct regression (or its nonparametric analogue), one must be willing to assume that true quality of performance influences salary over and above the influence of measures of performance. This involves postulating a latent variable, often labeled "qualifications," that has a direct effect on salary. For several reasons it may be undesirable to make this assumption. Other problems with the regression approach, however, do require serious consideration. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A method for detecting hydrophobic patches on protein surfaces

A method for the detection of hydrophobic patches on the surfaces of protein tertiary structures is presented. It delineates explicit contiguous pieces of surface of arbitrary size and shape that consist solely of carbon and sulphur atoms using a dot representation of the solvent-accessible surface. The technique is also useful in detecting surface segments with other characteristics, such as polar patches. Its potential as a tool in the study of protein-protein interactions and substrate recognition is demonstrated by applying the method to myoglobin, Leu/IIe/Val-binding protein, lipase, lysozyme, azurin, triose phosphate isomerase, carbonic anhydrase, and phosphoglycerate kinase. Only the largest patches, having sizes exceeding random expectation, are deemed meaningful. In addition to well-known hydrophobic patches on these proteins, a number of other patches are found, and their significance is discussed. The method is simple, fast, and robust. The program text is obtainable by anonymous ftp.     

Markov models of breast tumor progression: some age-specific results

Researchers have noted that mammographic screening has a reduced effect on breast cancer mortality in women in their forties compared to older women. Explanations for this include poorer sensitivity in younger women due to denser breast tissue, as well as more rapid tumor progression, giving a shorter mean sojourn time (the average duration of the preclinical screen-detectable period). To test these hypotheses, we developed a series of Markov-chain models to estimate tumor progression rates and sensitivity. Parameters were estimated using tumor data from the Swedish two-county trial of mammographic screening for breast cancer. The mean sojourn time was shorter in women aged 40-49 compared to women aged 50-59 and 60-69 (2.44, 3.70, and 4.17 years, respectively). Sensitivity was lower in the 40-49 age group compared to the two older groups (83%, 100%, and 100%, respectively). Thus, both rapid progression and poorer sensitivity are associated with the 40-49 age group. We also modeled tumor size, node status, and malignancy grade together with subsequent breast cancer mortality and found that, to achieve a reduction in mortality commensurate with that in women over 50, the interscreening interval for women in their forties should be less than two years. We conclude that Markov models and the use of tumor size, node status, and malignancy grade as surrogates for mortality can be useful in design and analysis of future studies of breast cancer screening.     

Enzyme-linked immunosorbent assay, using staphylococcal protein A for detecting virus antibodies

A modification of the indirect enzyme-linked immunosorbent assay (ELISA) was developed which used staphylococcal protein A linked to horseradish peroxidase. Virus antibodies in equine, bovine, porcine, feline, canine, lagomorphic (rabbit), and human sera were detected, using the indirect ELISA in which the antiglobulin enzyme conjugate was replaced by protein A linked to horseradish peroxidase. Results of the ELISA were compared with the results of the serum-virus neutralization test. The application of the test in laboratories performing serologic assays with sera from diverse animal species is discussed.     

Clinical applications of detecting dysfunctional p53 tumor suppressor protein

The p53 gene encodes for a protein, p53, which plays a critical role in controlling the cell cycle, in DNA repair and in programmed cell death (apoptosis). p53 is one of the most frequently mutated genes in human neoplasms and a variety of techniques have been developed to detect these mutations. These range from advanced molecular-genetic analyses to immunohistochemical staining for the p53 protein. This review will summarize our current understanding of the function of p53 as well as current methods to detect dysfunctional p53 and the clinical value of such analyses.     

Predicting peptides that bind to MHC molecules using supervised learning of hidden Markov models

Cells in the lateral hypothalamus and in the arcuate nucleus play prominent roles in the central control of food intake; however, a neurochemical link connecting these potential components of a hypothalamic circuitry regulating energy metabolism remains to be established. In the present study, the topographical relationship between cells expressing mRNAs encoding melanin-concentrating hormone and the newly discovered neuropeptide family hypocretins/orexins was studied in the rat and mouse lateral hypothalamus by using double-labeling in situ hybridization. Cells expressing the two mRNAs formed completely distinct populations, with hypocretin/orexin cells located primarily perifornically and in the magnocellular lateral hypothalamic nucleus; melanin-concentrating hormone cells extended in a wider area both laterally and periventricularly and appeared to partly surround the hypocretin/orexin population. In the arcuate nucleus, cells expressing neuropeptide Y and agouti gene-related protein were studied by routine fluorescence and/or confocal microscopy immunohistochemistry. Double staining demonstrated that a large proportion of the neuropeptide Y-positive cell bodies in this nucleus also contained agouti gene-related protein-like immunoreactivity. Moreover, these two peptides also coexisted in nerve terminals surrounding and in close relationship to perikarya and processes of both hypocretin/orexin- and melanin-concentrating hormone-immunoreactive cells in the lateral hypothalamus, whereby the former appeared to receive a more dense innervation. These results thus provide evidence for an arcuate-lateral hypothalamic neuropeptide Y/agouti gene-related protein pathway. Furthermore, the results implicate hypocretin/orexin and melanin-concentrating hormone-expressing cells as downstream targets in neuropeptide Y-induced feeding.     

Evaluation of protein A and protein G as indicator system in an ELISA for detecting antibodies in mink to Pseudomonas aeruginosa

A modified, indirect enzyme-linked immunosorbent assay (ELISA) was developed and applied in the detection of mink antibodies to Pseudomonas aeruginosa. In this assay, peroxidase conjugated protein A and protein G were evaluated as indicator systems for detecting antigen-antibody complexes. It was found that protein A has a strong affinity for mink immunoglobulins. In contrast, protein G showed no such affinity. The affinity of protein A for mink immunoglobulins was further demonstrated by immunoprecipitation assays.     

Validation of protein models derived from experiment

The growing number of protein structures solved at atomic resolution holds the promise of further improvements in geometry-based validation parameters. Additionally, the estimated standard uncertainties of the atomic coordinates have been computed for a number of X-ray structures, providing a measure of the coordinate precision. In NMR spectroscopy, a measure analogous to the crystallographic R-factor has been developed.     

Reduced protein models and their application to the protein folding problem

It has been well-recognized that starvation in anorexia and bulimia nervosa causes endocrine disturbances. Such disturbances may help understand why many people with eating disorders cannot easily reverse their illness since people with eating disorders often enter a downward spiraling circle with malnutrition sustaining and perpetuating the desire for more weight loss and dieting. Symptoms, such as obsessions and dysphoric mood, and altered appetitive behavior, may be exaggerated by neuropeptide alterations and thus contribute to this downward spiral. While neuropeptide disturbances do not appear to be a permanent feature or cause or anorexia nervosa, these disturbances are strongly entrenched, and are not easily corrected by improved nutrition or short-term weight normalization. This suggests that therapy should be sustained for months after nutritional normalization.     

Rational automatic search method for stable docking models of protein and ligand

An efficient automatic method has been developed for docking a ligand molecule to a protein molecule. The method can construct energetically favorable docking models, considering specific interactions between the two molecules and conformational flexibility in the ligand. In the first stage of docking, likely binding modes are searched and estimated effectively in terms of hydrogen bonds, together with conformations in part of the ligand structure that includes hydrogen bonding groups. After that part is placed in the protein cavity and is optimized, conformations in the remaining part are also examined systematically. Finally, several stable docking models are obtained after optimization of the position, orientation and conformation of the whole ligand molecule. In all the screening processes, the total potential energy including intra- and intermolecular interaction energy, consisting of van der Waals, electrostatic and hydrogen bonding energies, is used as the index. The characteristics of our docking method are high accuracy of the results, fully automatic generation of models and short computational time. The efficiency of the method was confirmed by four docking trials using two enzyme systems. In two attempts to dock methotrexate to dihydrofolate reductase and 2'-GMP to ribonuclease T1, the exact structures of complexes in crystals were reproduced as the most stable docking models, without any assumptions concerning the binding modes and ligand conformations. The most stable docking models of dihydrofolate and trimethoprim, respectively, to dihydrofolate reductase were also in good agreement with those suggested by experiment. In all test cases, it was shown that our method can accurately predict the correct docking structures, discriminating the correct model from incorrect ones. The efficiency of our method was further tested from the viewpoint of ability to predict the relative stability of the docking structures of two triazine derivatives to dihydrofolate reductase. Our docking method provides a useful tool for rational drug design and investigations of biochemical reaction mechanisms.