Mining Gene Expression Profile with Missing Values: An Integration of Kernel PCA and Robust Singular Values Decomposition

2018 ◽  
Vol 14 (1) ◽  
pp. 78-89 ◽  
Author(s):  
Md. Saimul Islam ◽  
Md. Aminul Hoque ◽  
Md. Sahidul Islam ◽  
Mohammad Ali ◽  
Md. Bipul Hossen ◽  
...  
Keyword(s):  
Gene Expression ◽  
Colon Cancer ◽  
Gene Expression Profiling ◽  
Microarray Data ◽  
Expression Profiling ◽  
Missing Values ◽  
Cellular Systems ◽  
Differentially Expressed ◽  
Data Matrix ◽  
Imputation Methods

Background: Gene expression profiling and transcriptomics provide valuable information about the role of genes that are differentially expressed between two or more samples. It is always important and challenging to analyse High-throughput DNA microarray data with a number of missing values under various experimental conditions. </P><P> Objectives: Graphical data visualizations of the expression of all genes in a particular cell provide holistic views of gene expression patterns, which improve our understanding of cellular systems under normal and pathological conditions. However, current visualization methods are sensitive to missing values, which are frequently observed in microarray-based gene expression profiling, potentially affecting the subsequent statistical analyses. Methods: We addressed in this study the problem of missing values with respect to different imputation methods using gene expression biplot (GE biplot), one of the most popular gene visualization techniques. The effects of missing values for mining differentially expressed genes in gene expression data were evaluated using four well-known imputation methods: Robust Singular Value Decomposition (Robust SVD), Column Average (CA), Column Median (CM), and K-nearest Neighbors (KNN). Frobenius norm and absolute distances were used to measure the accuracy of the methods. Results: Three numerical experiments were performed using simulated data (i) and publicly available colon cancer (ii) and leukemia data (iii) to analyze the performance of each method. The results showed that CM and KNN performed better than Robust SVD and CA for identifying the index gene profile in the biplot visualization in both the simulation study and the colon cancer and leukemia microarray datasets. Conclusion: The impact of missing values on the GE biplot was smaller when the data matrix was imputed by KNN than by CM. This study concluded that KNN performed satisfactorily in generating a GE biplot in the presence of missing values in microarray data.

scholarly journals Prognosis of stage II colon cancer by non-neoplastic mucosa gene expression profiling

Oncogene ◽  
2006 ◽  
Vol 26 (18) ◽  
pp. 2642-2648 ◽  
Author(s):  
A Barrier ◽  
F Roser ◽  
P-Y Boëlle ◽  
B Franc ◽  
C Tse ◽  
...  
Keyword(s):  
Gene Expression ◽  
Colon Cancer ◽  
Gene Expression Profiling ◽  
Expression Profiling ◽  
Stage Ii ◽  
Stage Ii Colon Cancer

scholarly journals Transcriptome analysis reveals that exogenous ethylene activates immune and defense responses in a high late blight resistant potato genotype

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Xiaohui Yang ◽  
Li Chen ◽  
Yu Yang ◽  
Xiao Guo ◽  
Guangxia Chen ◽  
...  
Keyword(s):  
Gene Expression ◽  
Late Blight ◽  
Gene Expression Profiling ◽  
Expression Profiling ◽  
Defense Responses ◽  
Differentially Expressed ◽  
Metabolic Pathway Analysis ◽  
Transcriptional Responses ◽  
Signaling Transduction Pathway ◽  
Potato Genotype

AbstractEthylene (ET) is one of the many important signaling hormones that functions in regulating defense responses in plants. Gene expression profiling was conducted under exogenous ET application in the high late blight resistant potato genotype SD20 and the specific transcriptional responses to exogenous ET in SD20 were revealed. Analysis of differentially expressed genes (DEGs) generated a total of 1226 ET-specific DEGs, among which transcription factors, kinases, defense enzymes and disease resistance-related genes were significantly differentially expressed. GO enrichment and KEGG metabolic pathway analysis also revealed that numerous defense regulation-related genes and defense pathways were significantly enriched. These results were consistent with the interaction of SD20 and Phytophthora infestans in our previous study, indicating that exogenous ET stimulated the defense response and initiated a similar defense pathway compared to pathogen infection in SD20. Moreover, multiple signaling pathways including ET, salicylic acid, jasmonic acid, abscisic acid, auxin, cytokinin and gibberellin were involved in the response to exogenous ET, which indicates that many plant hormones work together to form a complex network to resist external stimuli in SD20. ET-induced gene expression profiling provides insights into the ET signaling transduction pathway and its potential mechanisms in disease defense systems in potato.

scholarly journals Digital Gene Expression Profiling Analysis of Aged Mice under Moxibustion Treatment

2018 ◽  
Vol 2018 ◽  
pp. 1-10 ◽  
Author(s):  
Nan Liu ◽  
Yunyao Jiang ◽  
Min Xing ◽  
Baixiao Zhao ◽  
Jincai Hou ◽  
...  
Keyword(s):  
Gene Expression ◽  
Gene Ontology ◽  
Gene Expression Profiling ◽  
Differentially Expressed Genes ◽  
Pathway Analysis ◽  
Expression Profiling ◽  
Digital Gene Expression ◽  
Differentially Expressed ◽  
Aged Mice ◽  
Digital Gene Expression Profiling

Aging is closely connected with death, progressive physiological decline, and increased risk of diseases, such as cancer, arteriosclerosis, heart disease, hypertension, and neurodegenerative diseases. It is reported that moxibustion can treat more than 300 kinds of diseases including aging related problems and can improve immune function and physiological functions. The digital gene expression profiling of aged mice with or without moxibustion treatment was investigated and the mechanisms of moxibustion in aged mice were speculated by gene ontology and pathway analysis in the study. Almost 145 million raw reads were obtained by digital gene expression analysis and about 140 million (96.55%) were clean reads. Five differentially expressed genes with an adjusted P value < 0.05 and |log⁡2(fold  change)| > 1 were identified between the control and moxibustion groups. They were Gm6563, Gm8116, Rps26-ps1, Nat8f4, and Igkv3-12. Gene ontology analysis was carried out by the GOseq R package and functional annotations of the differentially expressed genes related to translation, mRNA export from nucleus, mRNA transport, nuclear body, acetyltransferase activity, and so on. Kyoto Encyclopedia of Genes and Genomes database was used for pathway analysis and ribosome was the most significantly enriched pathway term.

A New Molecular Classification of Multiple Myeloma Using Gene Expression Profiling and Fluorescence In Situ Hybridisation as Predictor for Event Free Survival.

Blood ◽  
2004 ◽  
Vol 104 (11) ◽  
pp. 73-73 ◽  
Author(s):  
Dirk Hose ◽  
Jean-Francois Rossi ◽  
Carina Ittrich ◽  
John deVos ◽  
Axel Benner ◽  
...  
Keyword(s):  
Gene Expression ◽  
Multiple Myeloma ◽  
Gene Expression Profiling ◽  
Expression Profiling ◽  
Molecular Classification ◽  
Differentially Expressed ◽  
Event Free Survival ◽  
Free Survival ◽  
Cytogenetic Aberrations

Abstract AIM was to establish a new molecular classification of Multiple Myeloma (MM) based on changes in global gene expression attributable to cytogenetic aberrations detected by interphase FISH (iFISH) in order to (i) predict event free survival (EFS) and (ii) investigate differentially expressed genes as basis for a group specific and risk adapted therapy. PATIENTS AND METHODS. Bone marrow aspirates of 105 newly diagnosed MM-patients (65 trial (TG) / 40 independent validation group (VG)) and 7 normal donors (ND) were CD138-purified by magnetic activated cell sorting. RNA was in-vitro transcribed and hybridised to Affymetrix HG U133 A+B GeneChip (TG) and HG U133 2.0 plus arrays (VG). CCND1 and CCND2 expression was verified by real time RT-PCR. iFISH was performed on purified MM-cells using probes for chromosomes 11q23, 11q13, 13q14, 17p13 and the IgH-translocations t(4;14) and t(11;14). Expression data were normalised (Bioconductor package gcrma) and nearest shrunken centroids (NSC) applied to calculate and cross validate a predictor on 40 patients of the TG with a comprehensive iFISH panel available combined with CCND overexpression. Differentially expressed genes were identified using empirical Bayes statistics for pairwise comparison. RESULTS. Overexpression of a D-type cyclin (D1 or D2) was found in 61/65 patients with MM compared to ND. CCND3 overexpression only appeared concomitantly with CCND2 overexpression. Four groups could be distinguished: (1.1) CCND1 (11q13) overexpression and trisomy 11q13, (1.2) CCND1 overexpression and translocations involving 11q13 i.e. t(11;14), (2.1) CCND2 overexpression without 11q13+, t(11;14), t(4;14), (2.2) CCND2 overexpression with t(4;14) and FGFR3 upregulation. A predictor of 6 to 566 genes correctly classifies all 40 patients of the TG (estimated cross validated error rate 0%). An independent VG of 40 patients was used. Genes with highest scores in NSC are: (1.1) CCND1, ribosomal proteins (e.g. RPL 28, 29), GPX1, CCRL2, (1.2) CCND1, TGIF, and NCAM (non-overexpression), (2.1) CCND2, (2.2) FGFR3, WHSC1, CCND2, IRTA2, SELL, and MAGED4. Distribution of clinical parameters (i.e. β2M, Durie Salmon stages, ISS) was not significantly different between the groups. The distribution of del(13)(q14q14) was (1.1) 31.5%, (1.2) 37.5%, (2.1) 37.5% and (2.2) 100%. (p<0.01). I.e. HGF, DKK1, VCAM, CD163 are differentially expressed between all 4 groups and ND (adjusted p<0.001). The groups defined by the predictor show a significantly different EFS after autologous stem cell transplantation according to the GMMG-HD3 protocol (median: (1.1) 18 / (1.2) not reached (no event) / (2.1) 22 / (2.2) 6 months; log-rank-test: p=0.004). CONCLUSION. CCND1 or CCND2 overexpression is nearly ubiquitous in MM and attributable to defined cytogenetic aberrations. Gene expression and iFISH allow a molecular classification of MM which can be predicted by gene expression profiling alone. Groups in the classification show a distinctive pattern in gene expression as well as a different EFS interpretable as risk stratification and indicator of therapeutic targets.

Gene Expression Profiling of Isolated Mesenchymal and Osteoblastic Cells Exhibits a Different Pattern of Expression in Multiple Myeloma Patients as Compared to Healthy Subjects: Potential Relationship with the Presence of Bone Lesions.

Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 3513-3513
Author(s):  
Nicola Giuliani ◽  
Katia Todoerti ◽  
Gina Lisignoli ◽  
Sara Tagliaferri ◽  
Luca Agnelli ◽  
...  
Keyword(s):  
Gene Expression ◽  
Multiple Myeloma ◽  
Cell Proliferation ◽  
Osteogenic Differentiation ◽  
Gene Expression Profiling ◽  
Expression Profiling ◽  
Differentially Expressed ◽  
Bone Lesions ◽  
Bone Status ◽  
Probe Set

Abstract Gene expression alterations occurring in the bone microenvironment cells and their potential relationships with the occurrence of bone lesions in multiple myeloma (MM) patients have never been investigated. In this study, we have isolated both mesenchymal (MSC) and osteoblastic (OB) cells, without in vitro differentiation, from bone biopsies obtained by iliac crest of 24 MM patients, 7 MGUS subjects and 8 healthy donors (N) who underwent orthopedics surgery. Bone status was evaluated in all MM patients by total X rays scan and MRI for the spine. Firstly, we evaluated cell proliferation in relationship with growth substrate (bone and glass) and cell phenotype by flow cytometry and immunohistochemistry. We found that both MSC and OB cells have higher cell doubling rate in MM patients as compared to N. Higher expression of alkaline phosphatase and Runx2 was observed in OB as compared to MSC cells in both N and MM patients without osteolytic lesions, but not in osteolytic ones. We performed a gene expression profiling analysis of isolated MSC and OB cells using GeneChip® Affymetrix HG-U133A oligonucleotide arrays. An unsupervised analysis of the most variable genes across the dataset generated a hierarchical clustering with the two major branches containing respectively MSC and OB samples. A multiclass analysis of N, MGUS and MM patients identified 33 differentially expressed probe-set (specific for 27 genes) in MSC cells, and 19 differentially expressed probe-set (13 genes) in OB, and the identified transcripts mainly characterized N versus MM and MGUS samples. A supervised analysis between N and MM samples identified 65 probes (56 genes: 17 up-regulated and 39 down-regulated) differentially expressed in MSC and 35 probes (29 genes, 12 up-regulated and 17 down-regulated) in OB. Notably, genes encoding the Homeobox class proteins, such as HOXB2-6-7, were up-regulated in both MSC and OB of MM patients as compared to N. As regards the bone status, a total of 60 probe-sets (3 up-regulated and 57 down-regulated genes) were found differentially expressed in MSC from osteolytic vs. non-osteolytic MM patients, whereas MGUS-MSC exhibited an intermediate transcriptional profile between osteolytic and non-osteolytic MM patients. A distinct pattern of gene expression profiling was also observed in MSC versus OB when osteolytic and non-osteolytic MM patients were compared (26 vs. 94 differentially expressed probe-sets, respectively), including transcription factors related to MSC osteogenic differentiation belonging to Runx2 pathway (HEY1) or Wnt and BMP signaling On the other hand, few genes were found differentially expressed in OB cells in relationship with the presence of bone lesions. In conclusion, we identified a distinctive transcriptional fingerprint in isolated MSC and OB cells of MM patients as compared to N subjects, which mainly correlated with cell proliferation. Moreover, a different gene expression profile was observed in MSC cells of MM patients according to the presence/absence of bone lesions, highlighting the critical role of the block of the osteogenic differentiation.

Gene expression profiling to predict recurrent Dukes' B colon cancer

2004 ◽  
Vol 22 (14_suppl) ◽  
pp. 9555-9555
Author(s):  
Y. Wang ◽  
T. Jatkoe ◽  
Y. Zhang ◽  
M. G. Mutch ◽  
D. Talantov ◽  
...  
Keyword(s):  
Gene Expression ◽  
Colon Cancer ◽  
Gene Expression Profiling ◽  
Expression Profiling
Sign in / Sign up

Export Citation Format

Share Document