PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 11532215

Cluster-Rasch models for microarray gene expression data.

H Li·F Hong
View original source ↗Export metadata ↓

Abstract

BACKGROUND: We propose two different formulations of the Rasch statistical models to the problem of relating gene expression profiles to the phenotypes. One formulation allows us to investigate whether a cluster of genes with similar expression profiles is related to the observed phenotypes; this model can also be used for future prediction. The other formulation provides an alternative way of identifying genes that are over- or underexpressed from their expression levels in tissue or cell samples of a given tissue or cell type. RESULTS: We illustrate the methods on available datasets of a classification of acute leukemias and of 60 cancer cell lines. For tumor classification, the results are comparable to those previously obtained. For the cancer cell lines dataset, we found four clusters of genes that are related to drug response for many of the 90 drugs that we considered. In addition, for each type of cell line, we identified genes that are over- or underexpressed relative to other genes. CONCLUSIONS: The cluster-Rasch model provides a probabilistic model for describing gene expression patterns across samples and can be used to relate gene expression profiles to phenotypes.

Explore related subjects

Acute DiseaseAntigens, NeoplasmBreast NeoplasmsCluster AnalysisDatabases, Nucleic AcidGene Expression ProfilingGene Expression RegulationHumansLeukemiaLeukemia, MyeloidModels, StatisticalOligonucleotide Array Sequence AnalysisPhenotypePrecursor Cell Lymphoblastic Leukemia-LymphomaProbabilityRegression AnalysisTumor Cells, CulturedJournal ArticleResearch Support, Non-U.S. Gov'tResearch Support, U.S. Gov't, P.H.S.

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

H Li, F Hong. 2001-07-31. Cluster-Rasch models for microarray gene expression data.. https://doi.org/10.1186/gb-2001-2-8-research0031

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2003

Laparoscopic cholecystectomy: are patients with biliary pancreatitis at increased operative risk?

BACKGROUND: Previous reports of laparoscopic cholecystectomy (LC) in patients with biliary pancreatitis suggested increased operative difficulty, high rates of conversion, and greater morbidity and mortality. METHODS: Between 1990 and 1997, LC was performed for biliary pancreatitis in 63 patients (Group I) and for other causes in 829 patients (Group II). RESULTS: Patients with biliary pancreatitis were significantly older (median age 57 vs 50 years, p = 0.009), with greater co-morbidity (ASA III/IV 24% vs 11%, p = 0.008). The groups were comparable with respect to the frequency of previous abdominal operations, acute inflammation of the gallbladder, and the frequency of bile duct calculi detected by intraoperative cholangiography. Moderate to severe adhesions involving the gallbladder were significantly more frequent in patients with biliary pancreatitis (46% vs 29%, p = 0.004). No significant differences were observed between the two groups with respect to intraoperative (1.5% Group I vs 6.0% Group II, p = 0.109) or postoperative complications (10% vs 8%, p = 0.426), conversion rate (0 vs 2.7%, p = 0.181), or duration of operation (median 92 vs 85 min, p = 0.33). CONCLUSION: Despite increased age and co-morbidity and more frequent adhesions, our data showed no evidence that intraoperative or postoperative complications were more frequent in patients with biliary pancreatitis than in other patients undergoing LC.

Acute Disease↗
PubMed2002

Membrane attack complex contributes to destruction of vascular integrity in acute lung allograft rejection.

The lung is known to be particularly susceptible to complement-mediated injury. Both C5a and the membrane attack complex (MAC), which is formed by the terminal components of complement (C5b-C9), can cause acute pulmonary distress in nontransplanted lungs. We used C6-deficient rats to investigate whether MAC causes injury to lung allografts. PVG.R8 lungs were transplanted orthotopically to MHC class I-incompatible PVG.1U recipients. Allografts from C6-sufficient (C6(+)) donors to C6(+) recipients were rejected with an intense vascular infiltration and diffuse alveolar hemorrhage 7 days after transplantation (n = 5). Ab and complement (C3d) deposition was accompanied by extensive vascular endothelial injury and intravascular release of von Willebrand factor. In contrast, lung allografts from C6-deficient (C6(-)) donors to C6(-) recipients survived 13-17 days (n = 5). In the absence of C6, perivascular mononuclear infiltrates of ED1(+) macrophages and CD8(+) T lymphocytes were present 7 days after transplantation, but vascular endothelial cells were quiescent, with minimal von Willebrand factor release and no evidence of alveolar hemorrhage or edema. Lung allografts were performed from C6(-) donors to C6(+) recipients (n = 5) and from C6(+) donors to C6(-) recipients (n = 5) to separate the effects of systemic and local C6 production. Lungs transplanted from C6(+) donors to C6(-) recipients had increased alveolar macrophages and capillary injury. C6 production by lung allografts was demonstrated at the mRNA and protein levels. These results demonstrate that MAC causes vascular injury in lung allografts and that the location of injury is dependent on the source of C6.

Acute Disease↗
PubMed2002

Dynamics of T cells and TCR excision circles differ after treatment of acute and chronic HIV infection.

We quantified T cell proliferation and thymic function in primary HIV infection (PHI; n = 19) and chronic HIV infection (CHI; n = 14) by measuring Ki67 staining and TCR excision circle (TREC) number. After antiretroviral therapy of PHI there is a profound decrease in the number and percentage of Ki67(+) T cells (<6% Ki67(+)) with no significant increase in TREC per million cells and a transient increase in TREC per milliliter. In contrast, after antiretroviral therapy of CHI there is a reduction in the percentage but little change in the total number of Ki67(+)CD4(+) T cells associated with increases in both TREC per million cells and TREC per milliliter. Using a mathematical model that accounts for proliferation, death, and redistribution of T cells, we find that redistribution is consistent with the TREC changes observed during treatment of PHI and that an increase in thymic output is needed to explain the increase in TREC during treatment of CHI. Consideration of TREC per milliliter shows that changes in proliferation alone cannot explain the changes in TREC. In addition, although increased proliferation of memory cells in HIV infection has been established, we find no difference in TREC per million CD45RA(-) "memory" T cells between healthy and infected individuals (p = 0.154 for CD4(+); p = 0.383 for CD8(+)). Finally, although the number of TREC per million cells is always much lower in memory T cells than in naive T cells, in the setting of HIV infection, given that memory cells make up a larger proportion of total T cells, we find that 50% of TREC per milliliter in CD4(+) T cells is harbored in the CD45RA(-) "memory" subset of our infected subjects.

Acute Disease↗