Functionally relevant prediction model for colorectal cancer
Ontology highlight
ABSTRACT:
ORGANISM(S): Homo sapiens
PROVIDER: GSE77434 | GEO | 2016/02/02
SECONDARY ACCESSION(S): PRJNA310328
REPOSITORIES: GEO
Similar Datasets
Project description:In this study, authors had briefly talked about the comparison of Support Vector Machines (SVM) and Artificial Neural Networks (ANN) for binary classification of cancer. It is concluded that, SVMs perform better on cancer prediction which are not expressed by their genotype whereas ANNs capture the flaws in the other case. In this paper, the author's had provided a good strategy to construct SVM model. The model was coded from their approach and Grid Search method was used for estimation of a better set of parameters. The resulted model had provided better evaluation metrics than the one mentioned in the manuscript. The model is then exported to Open Neural Network Exchange (ONNX) format to avail the model to be accessible in various platforms thereby promoting the FAIReR (Findable, Accessible, Interoperable, Reusable, and Reproducible) protocol for sharing machine learning models. Docker files were provided for both training and testing environment so that the curator can reproduce the results.
The prediction output '0' means the sample is Benign, otherwise its Malignant.
2024-07-13 | MODEL2407130001 | BioModels
Project description:Background and Aims: RNA biomarkers derived from sloughed enterocytes would provide an ideal, non-invasive method for early detection of colorectal cancer (CRC) and precancerous adenomas. To realize this goal, a highly reliable method to isolate preserved human RNA from stool samples is needed. Here we develop a protocol to identify RNA biomarkers associated with CRC to assess the use of these biomarkers for noninvasive screening of disease. Methods: Stool samples were collected from 454 patients prior to a colonoscopy. A nucleic acid extraction protocol was developed to isolate human RNA from 330 stool samples and transcript abundances were estimated by microarray analysis. This 330-patient cohort was split into a training set of 265 individuals to develop a machine learning model and a testing set of 65 individuals to determine the model’s ability to detect colorectal neoplasms. Results: Analysis of the transcriptome from 265 individuals identified 200 transcript clusters as differentially expressed (p<0.03). These transcripts were used to build a Support Vector Machine (SVM) based model to classify 65 individuals within the testing set. This SVM algorithm attained a 95% sensitivity for precancerous adenomas and a 65% sensitivity for CRC (stage I-IV). The machine learning algorithm attained a specificity of 59% for healthy individuals and an overall accuracy of 72.3%. Conclusions: We developed an RNA-based neoplasm detection model that is sensitive for CRC and precancerous adenomas. The model allows for non-invasive assessment of tumors and could potentially be used to provide clinical guidance for individuals within the screening population for colorectal cancer.
2019-02-25 | GSE99573 | GEO
Project description:Based on UPLC-Q-TOF-MS metabolomics technology, urine samples of 1072 subjects from 9 centers, including normal control, phlegm and blood stasis (PBS) syndrome and Qi and Yin deficiency (QYD) syndrome, and other syndromes of CHD, were conducted to find biomarkers. Among them, the discovery set (n = 125) and the test set (n = 337) were used to identify and validate biomarkers, and the validation set (n = 610) was used for the application and evaluation of the support vector machine (SVM) prediction model.
2019-03-27 | PXD012910 | 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:Ovarian cancer is the leading cause of death in gynaecological malignancies in women. However, currently there are no clinical or pathologic parameters available that can reliably predict clinical outcome. In a previously published pilot study we explored the performance of microarrays in predicting clinical behaviour of ovarian tumours. For this purpose we performed microarray analysis on 20 patients and estimated that we could predict disease stage with 100% accuracy and the response to platin-based chemotherapy with 76.92% accuracy using leave-one-out cross validation techniques in combination with Least Squares Support Vector Machines (LS-SVMs). In the current study we prospectively evaluate models, built on the pilot data set, on a set of 49 new patients. Principal component analysis showed that the gene expression data from stage I, platin-sensitive advanced stage and platin-resistant advanced stage tumours in the prospective study did not correspond to their respective classes in the pilot study. Additionally, LS-SVM models built using the data from the pilot study, only misclassified one of four stage I tumours and correctly classified all 45 advanced stage tumours but this model was not able to predict resistance to platin-based chemotherapy. We discuss possible reasons for prospective failure of these models and conclude that existing results based on gene expression patterns of ovarian tumours need to be thoroughly scrutinized before this technology could be considered ready for clinical use.
2007-12-31 | E-MEXP-995 | biostudies-arrayexpress
Project description:As systems biology approaches to virology have become more tractable, it has become possible to analyze highly studied viruses such as HIV in new, unbiased ways, including spatial proteomics. We have employed here a differential centrifugation protocol to fractionate an inducible model of HIV-expression in Jurkat T cells for proteomic analysis by mass spectrometry. Using these proteomics data, we evaluated the merits of several reported machine learning pipelines for classification of the spatial proteome and identification of protein translocations. From these analyses we found that classifier performance was organelle-dependent, with Bayesian t-augmented Gaussian mixture modeling outperforming support vector machine (SVM) learning for mitochondrial and ER proteins, but underperforming on cytosolic, nuclear, and plasma membrane proteins by QSep analysis. We also observed a generally higher performance for protein translocation identification using a Bayesian model, BANDLE, on SVM-classified data. Comparative BANDLE analysis of WT and ΔNef models also identified known Nef-dependent interactors such as TCR signaling and coatomer complex. Lastly, we found that SVM classification showed higher consistency and was less sensitive to HIV-dependent noise in our data. These findings illustrate important considerations for future studies of the spatial proteome following viral infection or expression where their generalizability can be further assessed.
2022-01-16 | PXD029956 | Pride
Project description:Psoriatic arthritis (PsA) is a chronic inflammatory joint disease, and the diagnosis is quite difficult due to the unavailability of reliable clinical markers. This study aimed to investigate the fecal metabolites in PsA by comparison with rheumatoid arthritis (RA), and to identify potential diagnostic biomarkers for PsA. The metabolic profiles of the fecal samples from 27 PsA and 29 RA patients as well as 36 healthy controls (HCs) were performed on ultra-high-performance liquid chromatography coupled with hybrid triple quadrupole time-of-flight mass spectrometry (UHPLC-Q-TOF-MS). And differentially altered metabolites were screened and assessed using multivariate analysis for exploring the potential biomarkers of PsA. The results showed that 154 fecal metabolites were significantly altered in PsA patients when compared with HCs, and 45 metabolites were different when compared with RA patients. A total of 14 common differential metabolites could be defined as candidate biomarkers. Furthermore, a support vector machines (SVM) model was performed to distinguish PsA from RA patients and HCs, and 5 fecal metabolites, including α/β-turmerone, glycerol 1-hexadecanoate, dihydrosphingosine, pantothenic acid and glutamine, were determined as biomarkers for PsA. Through the metabolic pathways analysis, we found that the abnormality of amino acid metabolism, bile acid metabolism and lipid metabolism might contribute to the occurrence and development of PsA. In summary, our research provided ideas for the early diagnosis and treatment of PsA by identifying fecal biomarkers and analyzing metabolic pathways.
2022-02-23 | MTBLS4206 | MetaboLights
Project description:Medullary breast cancers (MBC) display a basal profile, but a favorable prognosis. We hypothesized that a previously published 368-gene expression signature associated with MBC might serve to define a prognostic classifier in basal cancers. We collected public gene expression and histoclinical data of 2145 invasive early breast adenocarcinomas. We developed a Support Vector Machine (SVM) classifier based on this 368-gene list in a learning set, and tested its predictive performances in an independent validation set. Then, we assessed its prognostic value and that of six prognostic signatures for disease-free survival (DFS) in the remaining 2034 samples. The SVM model accurately classified all MBC samples in the learning and validation sets. A total of 466 cases were basal across other sets. The SVM classifier separated them into two subgroups, subgroup 1 (resembling MBC) and subgroup 2 (not resembling MBC). Subgroup 1 exhibited 71% 5-year DFS, whereas subgroup 2 exhibited 50% (p=9.93E-05). The classifier outperformed the classical prognostic variables in multivariate analysis, conferring lesser risk for relapse in subgroup 1 (HR=0.52, p=3.9E-04). This prognostic value was specific to the basal subtype, in which none of the other prognostic signatures was informative.
2010-05-05 | GSE21653 | GEO
Project description:Medullary breast cancers (MBC) display a basal profile, but a favorable prognosis. We hypothesized that a previously published 368-gene expression signature associated with MBC might serve to define a prognostic classifier in basal cancers. We collected public gene expression and histoclinical data of 2145 invasive early breast adenocarcinomas. We developed a Support Vector Machine (SVM) classifier based on this 368-gene list in a learning set, and tested its predictive performances in an independent validation set. Then, we assessed its prognostic value and that of six prognostic signatures for disease-free survival (DFS) in the remaining 2034 samples. The SVM model accurately classified all MBC samples in the learning and validation sets. A total of 466 cases were basal across other sets. The SVM classifier separated them into two subgroups, subgroup 1 (resembling MBC) and subgroup 2 (not resembling MBC). Subgroup 1 exhibited 71% 5-year DFS, whereas subgroup 2 exhibited 50% (p=9.93E-05). The classifier outperformed the classical prognostic variables in multivariate analysis, conferring lesser risk for relapse in subgroup 1 (HR=0.52, p=3.9E-04). This prognostic value was specific to the basal subtype, in which none of the other prognostic signatures was informative. The IPC series contained frozen tumor samples obtained from 266 early breast cancer patients who underwent initial surgery in our institution between 1992 and 2004. They included 227 cases previously reported {Finetti, 2008 #1758} and 39 additional cases, all similarly profiled using Affymetrix U133 Plus 2.0 human oligonucleotide microarrays as previously described {Finetti, 2008 #1758}. The study was approved by the IPC review board, and informed consent was available for each case. Gene expression data of 266 BCs were quantified by using whole-genome DNA microarrays (HG-U133 plus 2.0, Affymetrix).
2010-05-05 | E-GEOD-21653 | biostudies-arrayexpress
Project description:Background:
To assist clinicians with diagnosis and optimal treatment decision-making, we attempted to develop and validate an artificial intelligence prediction model for lung metastasis (LM) in colorectal cancer (CRC) patients.
Method:
The clinicopathological characteristics of 46037 CRC patients from the Surveillance, Epidemiology, and End Results (SEER) database and 2779 CRC patients from a multi-center external validation set were collected retrospectively. After feature selection by univariate and multivariate analyses, six machine learning (ML) models, including logistic regression, K-nearest neighbor, support vector machine, decision tree, random forest, and balanced random forest (BRF), were developed and validated for the LM prediction. The optimization model with best performance was compared to the clinical predictor. In addition, stratified LM patients by risk score were utilized for survival analysis.
| 46649 | ecrin-mdr-crc