Identification of biomarkers that distinguish chemical contaminants using a gradient feature selection method
Ontology highlight
ABSTRACT:
ORGANISM(S): Rattus norvegicus
PROVIDER: GSE19662 | GEO | 2010/01/05
SECONDARY ACCESSION(S): PRJNA122565
REPOSITORIES: GEO
Similar Datasets
Project description:There are many toxic chemicals to contaminate the world and cause harm to human and other organisms. How to quickly discriminate these compounds and characterize their potential molecular mechanism and toxicity is essential. High through put transcriptomics profiles such as microarray have been proven useful to identify biomarkers for different classification and toxicity prediction purposes. Here we aim to investigate how to use microarray to predict chemical contaminants and their possible mechanisms. In this study, we divided 105 compounds plus vehicle control into 14 compound classes. On the basis of gene expression profiles of in vitro primary cultured hepatocytes, we comprehensively compared various normalization, feature selection and classification algorithms for the classification of these 14 class compounds. We found that normalization had little effect on the averaged classification accuracy. Two support vector machine methods LibSVM and SMO had better classification performance. When feature sizes were smaller, LibSVM outperformed other classification methods. Simple logistic algorithm also performed well. At the training stage, usually the feature selection method SVM-RFE performed the best, and PCA was the poorest feature selection algorithm. But overall, SVM-RFE had the highest overfitting rate when an independent dataset used for a prediction in this case. Therefore, we developed a new feature selection algorithm called gradient method which had a pretty high training classification as well as prediction accuracy with the lowest over-fitting rate. Through the analysis of biomarkers that distinguished 14 class compounds, we found a goup of genes that mainly invovled in cell cylce were significanly downregulated by the metal and inflammatory compounds, but were induced by anti-microbial, cancer related drugs, pesticides, and PXR mediators. For in vitro experiment, primary cultured rat hepatocytes were treated one of 105 compounds with relative controls. At least three biological replicates were used for each unique condition. In total 531 arrays were used.
2010-01-05 | E-GEOD-19662 | biostudies-arrayexpress
Project description:Lung adenocarcinoma (LUAD) is one of the most common pathological and histological subtypes of primary lung cancer, with high morbidity and mortality. MicroRNAs (miRNAs) are endogenous small non-coding RNAs that regulate the expression of genes at post-transcriptional level. It was reported that A-to-I miRNA editing was decreased in tumors, suggesting the potential value of miRNA editing in cancer classification. However, existing miRNA-based cancer classification models mainly used the frequencies of miRNAs. In order to validate the contribution of miRNA editing information in cancer classification, we extracted three types of miRNA features, including the abundances of original miRNAs, the abundances of edited miRNAs, and the editing levels of miRNA editing sites. Our results show that four classification algorithms selected, i.e., kNN, C4.5, RF and SVM, generally had better performances on all features than on the abundances of miRNAs alone. Since the number of features were large, we used three feature selection (FS) methods to further improve the classification models. One of the FS methods, the DFL algorithm, selected only three features, i.e., the frequencies of hsa-miR-135b-5p, hsa-miR-210-3p and hsa-miR-182 48u (an edited miRNA), from 316 training samples. And all of the four classification algorithms achieved 100% accuracy on these three features for 79 independent testing samples. These results indicate that the additional information of miRNA editing are useful in improving the classification of LUAD samples. And the three miRNAs selected by DFL potentially represent an effective molecular signature for LUAD diagnosis.
2024-02-22 | GSE244311 | GEO
Project description:In an effort to develop a genomics-based approach to the prediction of drug response, we have developed an algorithm for classification of cell line chemosensitivity based on gene expression profiles alone. Using oligonucleotide microarrays, the expression levels of 6,817 genes were measured in a panel of 60 human cancer cell lines (the NCI-60) for which the chemosensitivity profiles of thousands of chemical compounds have been determined. We sought to determine whether the gene expression signatures of untreated cells were sufficient for the prediction of chemosensitivity. Gene expression-based classifiers of sensitivity or resistance for 232 compounds were generated and then evaluated on independent sets of data. The classifiers were designed to be independent of the cells' tissue of origin. The accuracy of chemosensitivity prediction was considerably better than would be expected by chance. Eighty-eight of 232 expression-based classifiers performed accurately (with P < 0.05) on an independent test set, whereas only 12 of the 232 would be expected to do so by chance. These results suggest that at least for a subset of compounds genomic approaches to chemosensitivity prediction are feasible.
2015-05-16 | GSE68936 | GEO
Project description:The well-defined battery of in vitro systems applied within chemical cancer risk assessment is often characterised by a high false-positive rate, thus repeatedly failing to correctly predict the in vivo genotoxic and carcinogenic properties of test compounds. Toxicogenomics, i.e. mRNA-profiling, has been proven successful in improving the prediction of genotoxicity in vivo and the understanding of underlying mechanisms. Recently, microRNAs have been discovered as post-transcriptional regulators of mRNAs. It is thus hypothesised that using microRNA response-patterns may further improve current prediction methods. This study aimed at predicting genotoxicity and non-genotoxic carcinogenicity in vivo, by comparing microRNA- and mRNA-based profiles, using a frequently applied in vitro liver model and exposing this to a range of well-chosen prototypical carcinogens. Primary mouse hepatocytes (PMH) were treated for 24 and 48h with 21 chemical compounds [genotoxins (GTX) vs. non-genotoxins (NGTX) and non-genotoxic carcinogens (NGTX-C) versus non-carcinogens (NC)]. MicroRNA and mRNA expression changes were analysed by means of Exiqon and Affymetrix microarray-platforms, respectively. Classification was performed by using Prediction Analysis for Microarrays (PAM). Compounds were randomly assigned to training and validation sets (repeated 10 times). Before prediction analysis, pre-selection of microRNAs and mRNAs was performed by using a leave-one-out t-test. No microRNAs could be identified that accurately predicted genotoxicity or non-genotoxic carcinogenicity in vivo. However, mRNAs could be detected which appeared reliable in predicting genotoxicity in vivo after 24h (7 genes) and 48h (2 genes) of exposure (accuracy: 90% and 93%, sensitivity: 65% and 75%, specificity: 100% and 100%). Tributylinoxide and para-Cresidine were misclassified. Also, mRNAs were identified capable of classifying NGTX-C after 24h (5 genes) as well as after 48h (3 genes) of treatment (accuracy: 78% and 88%, sensitivity: 83% and 83%, specificity: 75% and 93%). Wy-14,643, phenobarbital and ampicillin trihydrate were misclassified. We conclude that genotoxicity and non-genotoxic carcinogenicity probably cannot be accurately predicted based on microRNA profiles. Overall, transcript-based prediction analyses appeared to clearly outperform microRNA-based analyses.
2016-06-28 | GSE72014 | GEO
Project description:The well-defined battery of in vitro systems applied within chemical cancer risk assessment is often characterised by a high false-positive rate, thus repeatedly failing to correctly predict the in vivo genotoxic and carcinogenic properties of test compounds. Toxicogenomics, i.e. mRNA-profiling, has been proven successful in improving the prediction of genotoxicity in vivo and the understanding of underlying mechanisms. Recently, microRNAs have been discovered as post-transcriptional regulators of mRNAs. It is thus hypothesised that using microRNA response-patterns may further improve current prediction methods. This study aimed at predicting genotoxicity and non-genotoxic carcinogenicity in vivo, by comparing microRNA- and mRNA-based profiles, using a frequently applied in vitro liver model and exposing this to a range of well-chosen prototypical carcinogens. Primary mouse hepatocytes (PMH) were treated for 24 and 48h with 21 chemical compounds [genotoxins (GTX) vs. non-genotoxins (NGTX) and non-genotoxic carcinogens (NGTX-C) versus non-carcinogens (NC)]. MicroRNA and mRNA expression changes were analysed by means of Exiqon and Affymetrix microarray-platforms, respectively. Classification was performed by using Prediction Analysis for Microarrays (PAM). Compounds were randomly assigned to training and validation sets (repeated 10 times). Before prediction analysis, pre-selection of microRNAs and mRNAs was performed by using a leave-one-out t-test. No microRNAs could be identified that accurately predicted genotoxicity or non-genotoxic carcinogenicity in vivo. However, mRNAs could be detected which appeared reliable in predicting genotoxicity in vivo after 24h (7 genes) and 48h (2 genes) of exposure (accuracy: 90% and 93%, sensitivity: 65% and 75%, specificity: 100% and 100%). Tributylinoxide and para-Cresidine were misclassified. Also, mRNAs were identified capable of classifying NGTX-C after 24h (5 genes) as well as after 48h (3 genes) of treatment (accuracy: 78% and 88%, sensitivity: 83% and 83%, specificity: 75% and 93%). Wy-14,643, phenobarbital and ampicillin trihydrate were misclassified. We conclude that genotoxicity and non-genotoxic carcinogenicity probably cannot be accurately predicted based on microRNA profiles. Overall, transcript-based prediction analyses appeared to clearly outperform microRNA-based analyses.
2016-06-28 | GSE72081 | GEO
Project description:In an effort to develop a genomics-based approach to the prediction of drug response, we have developed an algorithm for classification of cell line chemosensitivity based on gene expression profiles alone. Using oligonucleotide microarrays, the expression levels of 6,817 genes were measured in a panel of 60 human cancer cell lines (the NCI-60) for which the chemosensitivity profiles of thousands of chemical compounds have been determined. We sought to determine whether the gene expression signatures of untreated cells were sufficient for the prediction of chemosensitivity. Gene expression-based classifiers of sensitivity or resistance for 232 compounds were generated and then evaluated on independent sets of data. The classifiers were designed to be independent of the cells' tissue of origin. The accuracy of chemosensitivity prediction was considerably better than would be expected by chance. Eighty-eight of 232 expression-based classifiers performed accurately (with P < 0.05) on an independent test set, whereas only 12 of the 232 would be expected to do so by chance. These results suggest that at least for a subset of compounds genomic approaches to chemosensitivity prediction are feasible. golub-00299 Assay Type: Gene Expression Provider: Affymetrix Array Designs: Hu6800 Organism: Homo sapiens (ncbitax) Tissue Sites: Kidney, Brain, Blood, Lung, Skin, Colon, Breast, Prostate, Ovary Material Types: cell, synthetic_RNA, whole_organism, total_RNA Disease States: Carcinoma, Glioblastoma, Lymphoma, Adenocarcinoma, Melanoma, Amelanotic Melanoma, Large Cell Carcinoma, Precursor Lymphoblastic Leukemia, Carcinosarcoma, Myeloid Leukemia, Plasma Cell Myeloma, Bronchogenic Carcinoma, Chronic Myelogenous Leukemia, Squamous Cell Carcinoma
2015-05-16 | E-GEOD-68936 | biostudies-arrayexpress
Project description:Background & Aims. The current staging system for colorectal cancer (CRC) based on TNM classification allows prediction of potential recurrence. However, it does not necessarily make reliable personalized prediction of prognosis. In this paper we describe combination of clinicopathological data and gene signature of dissected tumor specimen with stage II and III CRC patients would improve the situation.. Methods. A total of 1978 CRC were collected over 5 years, and then 371 stage II and 322 stage III of them with more than 45.9 months records were subjected to clinicopathological feature analyses. Out of this collection, 129 stage II and III CRC cases were selected for analyses of gene expression profiles with resected specimen. The gene signatures were analyzed by repeated random divisions of the samples into training and test sets to extract discriminator genes. After testing the applicability of this discriminator set, it was subjected to validation using a newly obtained set of 69 samples. Results. The pathological factors in solo or in combinations could not make personalized recurrence prediction, except for partial success with stage II patients. The gene signature, on the other hand, was capable of producing a set of discriminator genes, though the accuracy was yet to be improved. We observed that the best result was obtained when discriminators were selected from stage II CRC samples and used for prognosis of stage II CRC. When stage III cases were included in the process of discriminator extraction or in the process to validate samples, the results were poorer. Finally, we examined 31 independent stage II samples with a set of 30 such discriminators and were able to obtain results with 78 % accuracy, 90 % negative predictive value (NPV), and 55% positive predictive value (PPV). Conclusions. Independent clinicopathological variables were not able to predict prognosis of individual patient, unless the factors are combined. On the other hand, gene signatures allowed accurate prediction of prognosis for individuals, especially with stage II CRC, suggesting its potential use for selection of best treatment option for individual patients. The accuracy of discriminator prediction will be further improved when we take the evolution of CRC into consideration. Of 198 samples, 129 represented the discovery phase and 69 represented the validation phase.
2009-07-02 | E-GEOD-12032 | biostudies-arrayexpress
Project description:Systems-wide profiling of breast cancer has so far built on RNA and DNA analysis by microarray and sequencing techniques. Dramatic developments in proteomic technologies now enable very deep profiling of clinical samples, with high identification and quantification accuracy. We analyzed 40 estrogen receptor positive (luminal), Her2 positive and triple negative breast tumors and reached a quantitative depth of more than 10,000 proteins. Comparison to mRNA classifiers revealed multiple discrepancies between proteins and mRNA markers of breast cancer subtypes. These proteomic profiles identified functional differences between breast cancer subtypes, related to energy metabolism, cell growth, mRNA translation and cell-cell communication. Furthermore, we derived a 19-protein predictive signature, which discriminates between the breast cancer subtypes, through Support Vector Machine (SVM)-based classification and feature selection. The deep proteome profiles also revealed novel features of breast cancer subtypes, which may be the basis for future development of subtype specific therapeutics.
2016-01-05 | PXD002619 | Pride
Project description:Motivation: Monitoring, assessment and prediction of environmental risks that chemicals pose demand rapid and accurate diagnostic assays. A variety of toxicological effects have been associated with explosive compounds TNT and RDX. One important goal of microarray experiments is to discover novel biomarkers for toxicity evaluation. We have developed an earthworm microarray containing 15,208 unique oligo probes. Our objective was to identify biomarker genes that can separate earthworm samples into three groups: control, TNT-treated, and RDX-treated. Results: We developed a discriminant analysis and cluster (DAC) pipeline to analyze a 248-array dataset. First, a total of 869 significantly changed genes in response to TNT or RDX exposure were inferred by class comparison statistical algorithms. Then, nine decision tree-based algorithms were applied to generate classification rules and a set of 286 classifier genes. These classifier genes were ranked by their overall weight of significance, and were used to build support vector machines (SVMs). An SVM containing all 286 classifier genes had the highest classification accuracy (91.5%). An unsupervised clustering method was used to cluster the worm samples and results show that the use of the top 100 classifier genes can assign the largest number of worm samples into the three reference clusters obtained by using all the 14,188 filtered genes, suggesting that these top-ranked genes may be potential candidates for biomarkers. This study demonstrates that the DAC pipeline can be used to identify a small set of biomarker genes from high dimensional datasets and generate a reliable SVM classification model for multiple classes.
2010-10-08 | GSE18495 | GEO
Project description:Motivation: Monitoring, assessment and prediction of environmental risks that chemicals pose demand rapid and accurate diagnostic assays. A variety of toxicological effects have been associated with explosive compounds TNT and RDX. One important goal of microarray experiments is to discover novel biomarkers for toxicity evaluation. We have developed an earthworm microarray containing 15,208 unique oligo probes. Our objective was to identify biomarker genes that can separate earthworm samples into three groups: control, TNT-treated, and RDX-treated. Results: We developed a discriminant analysis and cluster (DAC) pipeline to analyze a 248-array dataset. First, a total of 869 significantly changed genes in response to TNT or RDX exposure were inferred by class comparison statistical algorithms. Then, nine decision tree-based algorithms were applied to generate classification rules and a set of 286 classifier genes. These classifier genes were ranked by their overall weight of significance, and were used to build support vector machines (SVMs). An SVM containing all 286 classifier genes had the highest classification accuracy (91.5%). An unsupervised clustering method was used to cluster the worm samples and results show that the use of the top 100 classifier genes can assign the largest number of worm samples into the three reference clusters obtained by using all the 14,188 filtered genes, suggesting that these top-ranked genes may be potential candidates for biomarkers. This study demonstrates that the DAC pipeline can be used to identify a small set of biomarker genes from high dimensional datasets and generate a reliable SVM classification model for multiple classes. Adult earthworms (E. fetida) were exposed in soil spiked with TNT (0, 6, 12, 24, 48, or 96 mg/kg) or RDX (8, 16, 32, 64, or 128 mg/kg) for 0, 4 or 14 days. The 4-day treatment was repeated with RDX concentration being 2, 4, 8, 16 or 32 mg/kg soil. Each treatment originally had 10 replicate worms with 8-10 survivors at the end of exposure. Total RNA was isolated from the surviving worms. A total of 248 worm RNA samples were hybridized to a custom-designed oligo array using Agilentâs one-color Low RNA Input Linear Amplification Kit. The array contains 15,208 non-redundant 60-mer probes, each targeting a unique E. fetida transcript (Gong et al. 2009). After hybridization and scanning, gene expression data were acquired using Agilentâs Feature Extraction Software (v.9.1.3). The 248-array dataset consists of three worm groups: 32 untreated controls, 96 TNT-treated, and 120 RDX-treated.
2010-10-08 | E-GEOD-18495 | biostudies-arrayexpress