Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “multiple clustering”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 919 records · Page 51Linked to original sources
PubMed2004

The lgtABCDE gene cluster, involved in lipooligosaccharide biosynthesis in Neisseria gonorrhoeae, contains multiple promoter sequences.

Biosynthesis of the variable core domain of lipooligosaccharide (LOS) in Neisseria gonorrhoeae is mediated by glycosyl transferases encoded by lgtABCDE. Changes within homopolymeric runs within lgtA, lgtC, and lgtD affect the expression state of these genes, with the nature of the LOS expressed determined by the functionality of these genes. However, the mechanism for modulating the amount of multiple LOS chemotypes expressed in a single cell is not understood. Using mutants containing polar disruptions within the lgtABCDE locus, we determined that the expression of this locus is mediated by multiple promoters and that disruption of transcription from these promoters alters the relative levels of simultaneously expressed LOS chemotypes. Expression of the lgtABCDE locus was quantified by using xylE transcriptional fusions, and the data indicate that this locus is transcribed in trace amounts and that subtle changes in transcription result in phenotypic changes. By using rapid amplification of 5' cDNA ends, transcriptional start sites and promoter sequences were identified within lgtABCDE. Most of these promoters possessed 50 to 67% homology with the consensus gearbox promoter sequence of Escherichia coli.

Bacterial Proteins↗
PubMed2006

Lactobacillus plantarum gene clusters encoding putative cell-surface protein complexes for carbohydrate utilization are conserved in specific gram-positive bacteria.

BACKGROUND: Genomes of gram-positive bacteria encode many putative cell-surface proteins, of which the majority has no known function. From the rapidly increasing number of available genome sequences it has become apparent that many cell-surface proteins are conserved, and frequently encoded in gene clusters or operons, suggesting common functions, and interactions of multiple components. RESULTS: A novel gene cluster encoding exclusively cell-surface proteins was identified, which is conserved in a subgroup of gram-positive bacteria. Each gene cluster generally has one copy of four new gene families called cscA, cscB, cscC and cscD. Clusters encoding these cell-surface proteins were found only in complete genomes of Lactobacillus plantarum, Lactobacillus sakei, Enterococcus faecalis, Listeria innocua, Listeria monocytogenes, Lactococcus lactis ssp lactis and Bacillus cereus and in incomplete genomes of L. lactis ssp cremoris, Lactobacillus casei, Enterococcus faecium, Pediococcus pentosaceus, Lactobacillius brevis, Oenococcus oeni, Leuconostoc mesenteroides, and Bacillus thuringiensis. These genes are neither present in the genomes of streptococci, staphylococci and clostridia, nor in the Lactobacillus acidophilus group, suggesting a niche-specific distribution, possibly relating to association with plants. All encoded proteins have a signal peptide for secretion by the Sec-dependent pathway, while some have cell-surface anchors, novel WxL domains, and putative domains for sugar binding and degradation. Transcriptome analysis in L. plantarum shows that the cscA-D genes are co-expressed, supporting their operon organization. Many gene clusters are significantly up-regulated in a glucose-grown, ccpA-mutant derivative of L. plantarum, suggesting catabolite control. This is supported by the presence of predicted CRE-sites upstream or inside the up-regulated cscA-D gene clusters. CONCLUSION: We propose that the CscA, CscB, CscC and CscD proteins form cell-surface protein complexes and play a role in carbon source acquisition. Primary occurrence in plant-associated gram-positive bacteria suggests a possible role in degradation and utilization of plant oligo- or poly-saccharides.

Antigens, Surface↗
PubMedDate not supplied

Assessing the predictive power of unsupervised visualization techniques to improve the identification of GPCR-focused compound libraries.

Principal component analysis and self-organizing maps (SOMs) were compared to cluster and visualize the chemical space of a large and diverse data set. The data set comprised about 3000 G-protein-coupled receptor (GPCR) ligands for about 130 receptors and 3000 non-GPCR ligands from the World Drug Index. The molecules were described with a topological pharmacophore point histogram descriptor and a chemical fingerprint descriptor. To assess the predictive power of the clustering, a leave-multiple-out cross validation with k nearest neighbor classification was performed. The results of the classification tests and the visualization showed a clear superiority of the SOM method. SOM correctly divided the data set into two main clusters, one for the GPCR and the other for the non-GPCR ligands. Our results suggest that a continuous GPCR-ligand space exists.

Drug Design↗
PubMedDate not supplied

Attribute clustering for grouping, selection, and classification of gene expression data.

This paper presents an attribute clustering method which is able to group genes based on their interdependence so as to mine meaningful patterns from the gene expression data. It can be used for gene grouping, selection, and classification. The partitioning of a relational table into attribute subgroups allows a small number of attributes within or across the groups to be selected for analysis. By clustering attributes, the search dimension of a data mining algorithm is reduced. The reduction of search dimension is especially important to data mining in gene expression data because such data typically consist of a huge number of genes (attributes) and a small number of gene expression profiles (tuples). Most data mining algorithms are typically developed and optimized to scale to the number of tuples instead of the number of attributes. The situation becomes even worse when the number of attributes overwhelms the number of tuples, in which case, the likelihood of reporting patterns that are actually irrelevant due to chances becomes rather high. It is for the aforementioned reasons that gene grouping and selection are important preprocessing steps for many data mining algorithms to be effective when applied to gene expression data. This paper defines the problem of attribute clustering and introduces a methodology to solving it. Our proposed method groups interdependent attributes into clusters by optimizing a criterion function derived from an information measure that reflects the interdependence between attributes. By applying our algorithm to gene expression data, meaningful clusters of genes are discovered. The grouping of genes based on attribute interdependence within group helps to capture different aspects of gene association patterns in each group. Significant genes selected from each group then contain useful information for gene expression classification and identification. To evaluate the performance of the proposed approach, we applied it to two well-known gene expression data sets and compared our results with those obtained by other methods. Our experiments show that the proposed method is able to find the meaningful clusters of genes. By selecting a subset of genes which have high multiple-interdependence with others within clusters, significant classification information can be obtained. Thus, a small pool of selected genes can be used to build classifiers with very high classification rate. From the pool, gene expressions of different categories can be identified.

Algorithms↗
PubMed2005

Molecular and chromosomal analysis of ribosomal cistrons in two cartilaginous fish, Taeniura lymma and Raja montagui (Chondrichthyes, Batoidea).

We used silver nitrate staining, CMA3 and FISH to study the chromosomal localization of both the major ribosomal genes and the nucleolar organizer regions as well as that of the minor ribosomal genes (5S rDNA) in two species of Batoidea, Taeniura lymma (Dasyatidae) and Raja montagui (Rajidae). In both species, all the metaphases examined showed the presence of multiple NOR-bearing sites, while the gene for 5S rRNA proved to be localized on two chromosome pairs. Furthermore, one of the two 5S rDNA sites in T. lymma was shown to be co-localized with the major ribosomal cluster. The presence of multiple nucleolar organizer regions in the two species might be interpreted as being the result of intraspecific polymorphisms, or as a phenomenon of the amplified transposition of mobile elements of the genome. We also determined the nucleotide sequence of the 5S rRNA gene, consisting of 564 bp in R. montagui and 612 bp in T. lymma. We also found TATA-like and (TGC)n trinucleotides, (CA)n dinucleotides and (GTGA)n tetranucleotides, which probably influence gene regulation.

Animals↗
PubMed2006

Spatial and temporal structure of typhoid outbreaks in Washington, D.C., 1906-1909: evaluating local clustering with the Gi* statistic.

BACKGROUND: To better understand the distribution of typhoid outbreaks in Washington, D.C., the U.S. Public Health Service (PHS) conducted four investigations of typhoid fever. These studies included maps of cases reported between 1 May - 31 October 1906 - 1909. These data were entered into a GIS database and analyzed using Ripley's K-function followed by the Gi* statistic in yearly intervals to evaluate spatial clustering, the scale of clustering, and the temporal stability of these clusters. RESULTS: The Ripley's K-function indicated no global spatial autocorrelation. The Gi* statistic indicated clustering of typhoid at multiple scales across the four year time period, refuting the conclusions drawn in all four PHS reports concerning the distribution of cases. While the PHS reports suggested an even distribution of the disease, this study quantified both areas of localized disease clustering, as well as mobile larger regions of clustering. Thus, indicating both highly localized and periodic generalized sources of infection within the city. CONCLUSION: The methodology applied in this study was useful for evaluating the spatial distribution and annual-level temporal patterns of typhoid outbreaks in Washington, D.C. from 1906 to 1909. While advanced spatial analyses of historical data sets must be interpreted with caution, this study does suggest that there is utility in these types of analyses and that they provide new insights into the urban patterns of typhoid outbreaks during the early part of the twentieth century.

Disease Outbreaks↗
PubMedDate not supplied

Are endothelial cell patterns of astrocytomas indicative of grade?

BACKGROUND: The most common primary brain tumors in children and adults are of astrocytic origin. Classic histologic grading schemes for astrocytomas have included evaluating the presence or absence of nuclear abnormalities, mitoses, vascular endothelial proliferation, and tumor necrosis. MATERIALS AND METHODS: We evaluated the vascular pattern of 17 astrocytoma surgical specimens (seven from children and 10 from adults), and four normal brains obtained at autopsy, utilizing antibody to glial fibrillary acidic protein (GFAP) and von Willebrand factor (vWF) utilizing confocal microscopy. A modified WHO classification was used. RESULTS: All tumor cases showed cells positive for GFAP. Control tissues showed a few, widely separated vessels. Pilocytic astrocytomas (four cases) showed lacy clusters of small-to-medium sized vessels, with intact vessel wall integrity. Diffuse, low grade astrocytoma (three cases) showed a staining pattern similar to control tissue; intermediate grade (one case), anaplastic astrocytoma (three cases) and gliobastoma multiforme (six cases) showed an increased vessel density with multiple small vessels (glomeruloid clusters), some with prominent intimal hyperplasia, loss of vessel wall integrity, and with numerous vWF-positive single cells/microvessels within the tumor substance. CONCLUSIONS: Evaluation of astrocytomas utilizing antibody to vWF and confocal microscopy aids in the grading of these neoplasms.

Adolescent↗
PubMed1986

Structure of perceptions of older adults: evidence for multiple stereotypes.

We examined the possibility that people hold multiple stereotypes of the elderly. Subjects were male and female university students. In the first phase of our study, stereotype content was sampled by asking subjects to describe the typical old person. In the second phase of the study, different subjects sorted traits from Phase 1 descriptions into one or more groups. Each group contained those traits that subjects felt could be found in one and the same older adult. Attitudes toward the stereotypes were also assessed. A distance matrix, based on the number of subjects who sorted each pair of traits into different groups, was analyzed by hierarchical cluster analysis. Evidence for multiple stereotypes was found both in the presence of contradictory traits given in Phase 1 descriptions and in the structure of the clusters. Different attitudes are identified for the cluster-defined stereotypes.

Adolescent↗
PubMed1999

Mechanisms of genomic imprinting.

A small number of mammalian genes undergo the process of genomic imprinting whereby the expression level of the alleles of a gene depends upon their parental origin. In the past year, attention has focused on the mechanisms that determine parental-specific expression patterns. Many imprinted genes are located in conserved clusters and, although it is apparent that imprinting of adjacent genes is jointly regulated, multiple mechanisms among and within clusters may operate. Recent developments have also refined the timing of the gametic imprints and further defined the mechanism by which DNA methyltransferases confer allelic methylation patterns.

Alleles↗
PubMed1984

Methylation of the hypoxanthine phosphoribosyltransferase locus on the human X chromosome: implications for X-chromosome inactivation.

To explore the role of DNA methylation in maintaining dosage compensation of X chromosome-linked genes and in regulating the transcriptional activity of "housekeeping" genes, we characterized DNA methylation of active, inactive, and derepressed alleles at the locus for hypoxanthine phosphoribosyltransferase (HPRT) on the human X chromosome. The methylation of Hpa II and Hha I sites in HPRT alleles on the active X chromosome was the same in all tissues. The consensus pattern includes hypomethylation of 5' clustered sites and extensive methylation of the 3' sequence. The striking feature of methylation of inactive X-chromosome alleles is nonuniformity and less extensive hypomethylation of the 5' cluster. Analysis of HPRT alleles reactivated in response to 5-azacytidine showed at least partial restoration of the consensus pattern. These observations indicate that methylation of housekeeping genes on the X chromosome is the same as that of autosomal ones and that the overall pattern and methylation of multiple sites within a cluster may cooperate to facilitate transcription. Furthermore, the fidelity of methylation of the active allele and the extensive drift in methylation of the inactive allele suggest that mechanisms involved in X-chromosome dosage compensation may be directed at the active rather than inactive X chromosome.

Alleles↗
PubMed2003

Carcinogen-induced impairment of enzymes for replicative fidelity of DNA and the initiation of tumours.

Not all carcinogens are mutagens, and many mutagens are not carcinogens. Among related chemicals, small changes of structure can markedly influence carcinogenic potency. Many tumours are genetically unstable, but some, especially 'benign' types, rarely exhibit 'progression' or show other evidence of genetic instability. Cells of particular tumour types exhibit identifiable particular 'sets' of phenotypic abnormalities (e.g. rapid growth, uniform nuclei, little cytoplasm and occasionally production of adrenocorticotrophic hormone by anaplastic small-celled carcinoma of the bronchus). Tumour cells pass their abnormalities on to their daughter cells, indicating that a genomic alteration probably underlies tumour formation. A possible mechanism, which might explain these phenomena is carcinogen-induced reduction of fidelity of replication of DNA polymerase complexes during S phase of normal tissue stem cells. A single 'hit' by a reactive agent (chemical or physical) on one of the major enzymic sites (synthesis, proofreading, mismatch repair-MMR) could cause multiple sequence abnormalities in the length of DNA synthesized by one DNA polymerase complex. Because this length of DNA (half a replication 'bubble') averages 15 000-150 000 nucleotides, the affected DNA could include two or more significant genomic elements (genes, especially for tumour suppression, regulatory loci and other elements). The particular mutant elements in the affected DNA could then determine the 'set' of phenotypic abnormalities exhibited by a resulting tumour. Non-genotoxic carcinogenicity, non-carcinogenic mutagenicity, structure-dependent chemical carcinogenicity and the phenomenon of 'sets' of phenotypic abnormalities could thus be accommodated. In experimental studies, the 'hallmark pattern' of mutation caused by this mechanism would be multiple mainly point mutations clustered within the length of half a replication 'bubble'. Such a 'hallmark pattern' of mutation might be detectable in carcinogen-treated cell cultures by the use of cycle-synchronized cultures, single cell subculturing, restriction (endonuclease) fragment length analysis of the clones and nucleotide sequencing of abnormal bands for localization in the human genome. If the mechanism is important to carcinogenesis generally, then non-carcinogenic mutagens should not cause the 'hallmark pattern' of mutations in either in vitro or in vivo systems. In human tumour cells, the 'hallmark pattern' of mutations may be demonstrable in genetically stable human tumours, but might well be lost or obscured by secondary mutations in genetically unstable tumours. Among different cases of the same type of human tumour, the clustered point mutations might be tumour-type specific in their location in the genome, but vary case-to-case in the precise 'points' mutated in the cluster region. New assays for assessing the carcinogenic potential of environmental and synthetic substances for human and animal populations may result. The hypothesis is not put forward to the exclusion of some established mechanisms of carcinogenesis for particular human tumours: for example, the 'two-hit' mutational hypothesis for retinoblastoma, the 'multiple sequential mutational' hypothesis for UV-induced lesions of the epidermis, and the possibility of adduct-induced frameshift mutations by some chemical carcinogens for experimental tumours.

Animals↗
PubMed2006

Simulation studies of the fidelity of biomolecular structure ensemble recreation.

We examine the ability of Bayesian methods to recreate structural ensembles for partially folded molecules from averaged data. Specifically we test the ability of various algorithms to recreate different transition state ensembles for folding proteins using a multiple replica simulation algorithm using input from "gold standard" reference ensembles that were first generated with a Go-like Hamiltonian having nonpairwise additive terms. A set of low resolution data, which function as the "experimental" phi values, were first constructed from this reference ensemble. The resulting phi values were then treated as one would treat laboratory experimental data and were used as input in the replica reconstruction algorithm. The resulting ensembles of structures obtained by the replica algorithm were compared to the gold standard reference ensemble, from which those "data" were, in fact, obtained. It is found that for a unimodal transition state ensemble with a low barrier, the multiple replica algorithm does recreate the reference ensemble fairly successfully when no experimental error is assumed. The Kolmogorov-Smirnov test as well as principal component analysis show that the overlap of the recovered and reference ensembles is significantly enhanced when multiple replicas are used. Reduction of the multiple replica ensembles by clustering successfully yields subensembles with close similarity to the reference ensembles. On the other hand, for a high barrier transition state with two distinct transition state ensembles, the single replica algorithm only samples a few structures of one of the reference ensemble basins. This is due to the fact that the phi values are intrinsically ensemble averaged quantities. The replica algorithm with multiple copies does sample both reference ensemble basins. In contrast to the single replica case, the multiple replicas are constrained to reproduce the average phi values, but allow fluctuations in phi for each individual copy. These fluctuations facilitate a more faithful sampling of the reference ensemble basins. Finally, we test how robustly the reconstruction algorithm can function by introducing errors in phi comparable in magnitude to those suggested by some authors. In this circumstance we observe that the chances of ensemble recovery with the replica algorithm are poor using a single replica, but are improved when multiple copies are used. A multimodal transition state ensemble, however, turns out to be more sensitive to large errors in phi (if appropriately gauged) and attempts at successful recreation of the reference ensemble with simple replica algorithms can fall short.

Algorithms↗
PubMed1997

[The role of insulin resistance in the clustering of cardiovascular risk factors].

The coexistence of multiple cardiovascular risk factors in the same individual is called a clustering of cardiovascular risk factors. Many studies have shown that this phenomenoa is associated with a increased morbility of cardiovascular disease, and insulin resistance may be the common link between the risk factors. This study was designed to link the clustering of cardiovascular risk factors with insulin resistance. Using a new insulin sensitivity index [-log (fasting serum Glu x Ins)] to evaluate the insulin resistance, we investigated the relationship between the insulin sensitivity and cardiovascular risk factors in 106 patients with hypertension or coronary heart disease and 32 normal subjects. The normal group without cardiovascular risk factors was selected as controls. The individual with 2 risk factors were defined as having mild clustering, and those with > or = 3 risk factors as severe clustering. As the number of risk factors increased from 1 to > or = 3, the insulin sensitivity index increased from -1.69 +/- 0.24, -1.95 +/- 0.17 to -2.14 +/- 0.21 respectively. The insulin resistance was positively correlated with serum uric acid in women (P < 0.05), but not in men (P > 0.05). The insulin sensitivity was more decreased in postmenopausal than in pre-menopausal women. The insulin sensitivity index was -2.23 +/- 0.39 vs -1.73 +/- 0.13 (P < 0.05). The insulin resistance is associated with a clustering of risk factors of cardiovascular disease, more cluste-ring, more resistant.

Adult↗
PubMed1998

Transcription of the amoC, amoA and amoB genes in Nitrosomonas europaea and Nitrosospira sp. NpAV.

Nitrifying bacteria such as Nitrosomonas europaea and Nitrosospira sp. NpAV use ammonia monooxygenase (AMO) for oxidation of their primary growth substrate, ammonia. Two polypeptides of AMO are coded for by contiguous genes, amoA and amoB, which are preceded by a third gene, amoC. The amoCAB clusters are present in multiple copies in nitrifying bacteria of the beta subdivision. These bacteria also have one amoC copy that is not adjacent to a copy of amoAB. The seven known amoC genes in different nitrifiers code for similar polypeptides (> 68%). Reverse transcriptase-polymerase chain reactions and Northern blots indicated that amoC from the amoCAB cluster is contained on a transcript with amoAB. Two other transcripts were detected with amo probes and may be a product of processing of the amoCAB mRNA or independent transcripts.

Amino Acid Sequence↗
PubMed2006

Reconstruction and regulation of the central catabolic pathway in the thermophilic propionate-oxidizing syntroph Pelotomaculum thermopropionicum.

Obligate anaerobic bacteria fermenting volatile fatty acids in syntrophic association with methanogenic archaea share the intermediate bottleneck step in organic-matter decomposition. These organisms (called syntrophs) are biologically significant in terms of their growth at the thermodynamic limit and are considered to be the ideal model to address bioenergetic concepts. We conducted genomic and proteomic analyses of the thermophilic propionate-oxidizing syntroph Pelotomaculum thermopropionicum to obtain the genetic basis for its central catabolic pathway. Draft sequencing and subsequent targeted gap closing identified all genes necessary for reconstructing its propionate-oxidizing pathway (i.e., methylmalonyl coenzyme A pathway). Characteristics of this pathway include the following. (i) The initial two steps are linked to later steps via transferases. (ii) Each of the last three steps can be catalyzed by two different types of enzymes. It was also revealed that many genes for the propionate-oxidizing pathway, except for those for propionate coenzyme A transferase and succinate dehydrogenase, were present in an operon-like cluster and accompanied by multiple promoter sequences and a putative gene for a transcriptional regulator. Proteomic analysis showed that enzymes in this pathway were up-regulated when grown on propionate; of these enzymes, regulation of fumarase was the most stringent. We discuss this tendency of expression regulation based on the genetic organization of the open reading frame cluster. Results suggest that fumarase is the central metabolic switch controlling the metabolic flow and energy conservation in this syntroph.

Acyl Coenzyme A↗
PubMed2002

Familial occurrence of autoimmune diseases and autoantibodies in a Caucasian population of patients with systemic lupus erythematosus.

To determine the prevalence of autoimmune diseases and autoantibodies in relatives of Caucasian patients with systemic lupus erythematosus (SLE) we questioned 118 patients for the prevalence of autoimmune diseases in their relatives. Multicase SLE families were selected for further investigation: assessment of the presence of antinuclear antibodies (ANA), thyroid antibodies and IgM rheumatoid factor (IgM-RF). Thirty-three patients reported the presence of 50 autoimmune diseases in 43 relatives. Twenty-two diagnoses could be either confirmed (n=14) or refuted (n=8). SLE clustered significantly within families of SLE patients. Multiple sclerosis and rheumatoid arthritis also seemed to cluster within families of lupus patients. The prevalence of ANA (24%) and thyroid antibodies (44%) in 29 relatives of multicase SLE families was raised (P<0.05). In conclusion, the prevalence of autoimmune diseases is raised in relatives of Caucasian SLE patients. Also, the prevalence of autoantibodies is raised in relatives of multicase SLE families, both suggesting genetic influences in the pathogenesis of the disease. These findings support the genome-wide screening of SLE patients to unravel factors responsible for genetic susceptibility to SLE.

Adult↗
PubMed2000

An analysis of 14 molecular markers for monitoring osteoarthritis: segregation of the markers into clusters and distinguishing osteoarthritis at baseline.

OBJECTIVE: To investigate the relationships between serum and urinary molecular markers (MM) used to monitor osteoarthritis. DESIGN: Forty osteoarthritis patients had blood and urine collected at baseline and 1, 3, 6 and 12 months later. Specimens from 20 controls were obtained twice at a one month interval. The concentration of 14 different markers was determined at each time point and the data were analyzed by statistical methodology. RESULTS: The markers could be divided by the method of principal components analysis into five clusters of related markers: inflammation markers (C-reactive protein, tumor necrosis receptor type I and tumor necrosis receptor type II, interleukin 6, eosinophilic cationic protein), bone markers (bone sialoprotein, hydroxylysyl pyridinoline, lysyl pyridinoline), putative markers of cartilage anabolism (carboxypropeptide of type II procollagen, hyaluronan, epitope 846) and catabolism (keratan sulfate, cartilage oligomeric matrix protein), and transforming growth factor beta. Three markers (tumor necrosis factor receptor II, cartilage oligomeric matrix protein and epitope 846) from independent clusters discriminated osteoarthritis patients from controls. Inflammation was not a confounding factor in measurement, but a recognizable distinguishing factor in osteoarthritis. CONCLUSIONS: The markers separated into rational groups on the basis of their covariance, a finding with independent biochemical support. The covariance of markers from the same cluster suggests the use of a representative marker from the cluster to reflect changes in osteoarthritis. If multiple markers are being measured within a single cluster, then the use of a weighted cluster 'factor' may be preferable to the separate use of individual markers.

Amino Acids↗
PubMed2005

A semiparametric mixture model for analyzing clustered competing risks data.

A very general class of multivariate life distributions is considered for analyzing failure time clustered data that are subject to censoring and multiple modes of failure. Conditional on cluster-specific quantities, the joint distribution of the failure time and event indicator can be expressed as a mixture of the distribution of time to failure due to a certain type (or specific cause), and the failure type distribution. We assume here the marginal probabilities of various failure types are logistic functions of some covariates. The cluster-specific quantities are subject to some unknown distribution that causes frailty. The unknown frailty distribution is modeled nonparametrically using a Dirichlet process. In such a semiparametric setup, a hybrid method of estimation is proposed based on the i.i.d. Weighted Chinese Restaurant algorithm that helps us generate observations from the predictive distribution of the frailty. The Monte Carlo ECM algorithm plays a vital role for obtaining the estimates of the parameters that assess the extent of the effects of the causal factors for failures of a certain type. A simulation study is conducted to study the consistency of our methodology. The proposed methodology is used to analyze a real data set on HIV infection of a cohort of female prostitutes in Senegal.

AIDS Vaccines↗