Prediction of Drug Efficiency by Transferring Gene Expression Data from Cell Lines to Cancer Patients

2018 ◽  
pp. 201-212 ◽  
Author(s):  
Nicolas Borisov ◽  
Victor Tkachev ◽  
Anton Buzdin ◽  
Ilya Muchnik
Keyword(s):  
Gene Expression ◽  
Cancer Patients ◽  
Cell Lines ◽  
Gene Expression Data ◽  
Expression Data ◽  
Drug Efficiency

scholarly journals A method of gene expression data transfer from cell lines to cancer patients for machine-learning prediction of drug efficiency

Cell Cycle ◽  
2018 ◽  
Vol 17 (4) ◽  
pp. 486-491 ◽  
Author(s):  
Nicolas Borisov ◽  
Victor Tkachev ◽  
Maria Suntsova ◽  
Olga Kovalchuk ◽  
Alex Zhavoronkov ◽  
...  
Keyword(s):  
Gene Expression ◽  
Machine Learning ◽  
Cancer Patients ◽  
Cell Lines ◽  
Gene Expression Data ◽  
Data Transfer ◽  
Expression Data ◽  
Drug Efficiency

scholarly journals GEDS: A Gene Expression Display Server for mRNAs, miRNAs and Proteins

Cells ◽  
2019 ◽  
Vol 8 (7) ◽  
pp. 675 ◽  
Author(s):  
Xia ◽  
Liu ◽  
Zhang ◽  
Guo
Keyword(s):  
Gene Expression ◽  
Cell Lines ◽  
Gene Expression Data ◽  
Expression Profiles ◽  
Gene Expression Profiles ◽  
Cancer Cell Line ◽  
Tissue Expression ◽  
The Cancer Genome Atlas ◽  
Expression Data ◽  
Protein Levels

High-throughput technologies generate a tremendous amount of expression data on mRNA, miRNA and protein levels. Mining and visualizing the large amount of expression data requires sophisticated computational skills. An easy to use and user-friendly web-server for the visualization of gene expression profiles could greatly facilitate data exploration and hypothesis generation for biologists. Here, we curated and normalized the gene expression data on mRNA, miRNA and protein levels in 23315, 9009 and 9244 samples, respectively, from 40 tissues (The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GETx)) and 1594 cell lines (Cancer Cell Line Encyclopedia (CCLE) and MD Anderson Cell Lines Project (MCLP)). Then, we constructed the Gene Expression Display Server (GEDS), a web-based tool for quantification, comparison and visualization of gene expression data. GEDS integrates multiscale expression data and provides multiple types of figures and tables to satisfy several kinds of user requirements. The comprehensive expression profiles plotted in the one-stop GEDS platform greatly facilitate experimental biologists utilizing big data for better experimental design and analysis. GEDS is freely available on http://bioinfo.life.hust.edu.cn/web/GEDS/.

scholarly journals Systematic assessment of multi-gene predictors of pan-cancer cell line sensitivity to drugs exploiting gene expression data

F1000Research ◽  
2016 ◽  
Vol 5 ◽  
pp. 2927 ◽  
Author(s):  
Linh Nguyen ◽  
Cuong C Dang ◽  
Pedro J. Ballester
Keyword(s):  
Gene Expression ◽  
Machine Learning ◽  
Cell Line ◽  
Cell Lines ◽  
Gene Expression Data ◽  
Single Gene ◽  
Cancer Cell Line ◽  
Expression Data ◽  
Gene Markers ◽  
Pan Cancer

Background:Selected gene mutations are routinely used to guide the selection of cancer drugs for a given patient tumour. Large pharmacogenomic data sets were introduced to discover more of these single-gene markers of drug sensitivity. Very recently, machine learning regression has been used to investigate how well cancer cell line sensitivity to drugs is predicted depending on the type of molecular profile. The latter has revealed that gene expression data is the most predictive profile in the pan-cancer setting. However, no study to date has exploited GDSC data to systematically compare the performance of machine learning models based on multi-gene expression data against that of widely-used single-gene markers based on genomics data.Methods:Here we present this systematic comparison using Random Forest (RF) classifiers exploiting the expression levels of 13,321 genes and an average of 501 tested cell lines per drug. To account for time-dependent batch effects in IC50measurements, we employ independent test sets generated with more recent GDSC data than that used to train the predictors and show that this is a more realistic validation than K-fold cross-validation.Results and Discussion:Across 127 GDSC drugs, our results show that the single-gene markers unveiled by the MANOVA analysis tend to achieve higher precision than these RF-based multi-gene models, at the cost of generally having a poor recall (i.e. correctly detecting only a small part of the cell lines sensitive to the drug). Regarding overall classification performance, about two thirds of the drugs are better predicted by multi-gene RF classifiers. Among the drugs with the most predictive of these models, we found pyrimethamine, sunitinib and 17-AAG.Conclusions:We now know that this type of models can predictin vitrotumour response to these drugs. These models can thus be further investigated onin vivotumour models.

Novel Epigenetic 12-gene Signature Predictive of Poor Prognosis and MSI-like Phenotype in Human Metastatic Colorectal Carcinomas

2020 ◽  
Author(s):  
Valentina Condelli ◽  
Giovanni Calice ◽  
Alessandra Cassano ◽  
Michele Basso ◽  
Maria Grazia Rodriquenz ◽  
...  
Keyword(s):  
Gene Expression ◽  
Cell Lines ◽  
Gene Expression Data ◽  
Poor Prognosis ◽  
Expression Profiles ◽  
Hct116 Cell ◽  
Gene Copy ◽  
Expression Data ◽  
Global Methylation ◽  
Differentially Methylated Genes

Abstract Background. Epigenetic remodeling is responsible for tumor progression and drug resistance in human colorectal carcinoma (CRC). A subgroup of human CRCs exhibits the CIMP status, with extensive hypermethylation events in promoter regions of several genes, even though the prognostic significance of CIMP is controversial. This study addressed the hypothesis that DNA methylation profiling may identify metastatic CRC (mCRC) subtypes with different clinical behavior. Methods. Global methylation profile was comparatively analyzed between 24 first-line primary-resistant and 12 drug-sensitive mCRCs (in-house cohort), two subgroups of tumors with significantly different outcome. Methylation and gene expression data from 33 mCRCs of the TCGA COAD dataset (TCGA COAD cohort) were used to identify, among differentially methylated genes, a prognostic signature of functionally methylated genes. Clusters of mCRCs with different methylation patterns were further characterized for DNA mutational load, gene copy number and gene expression profiles. Human CRC HT29 and HCT116 cell lines were adapted to growth in presence of oxaliplatin and irinotecan and used as in vitro model to validate gene expression data.Results. Twelve functionally methylated genes yielded a hierarchical clustering of patients in two well-defined clusters with hypermethylated tumors characterized by a significantly worse relapse-free and overall survival compared to hypomethylated cancers and this was reproduced in both the in-house and the TCGA COAD cohorts. Interestingly, the hypermethylated poor prognosis cluster was enriched of CIMP-high and MSI-like cases. Furthermore, methylation events were enriched in genes located on q-arm of chromosomes 13 and 20, two chromosomal regions with gain/loss alterations strongly associated with adenoma-to-carcinoma progression. Finally, the expression of the 12-genes signature and MSI-enriching genes was confirmed in two independent oxaliplatin- and irinotecan-resistant CRC cell lines. Conclusions. These data represent the proof of concept that the hypermethylation of specific sets of genes may provide prognostic information being able to identify a subgroup of mCRCs with poor prognosis.

scholarly journals Supervised Classification by Filter Methods and Recursive Feature Elimination Predicts Risk of Radiotherapy-Related Fatigue in Patients with Prostate Cancer

2014 ◽  
Vol 13 ◽  
pp. CIN.S19745 ◽  
Author(s):  
Leorey N. Saligan ◽  
Juan Luis Fernández-Martínez ◽  
Enrique J. deAndrés-Galiana ◽  
Stephen Sonis
Keyword(s):  
Gene Expression ◽  
Prostate Cancer ◽  
Cancer Patients ◽  
Gene Expression Data ◽  
Fold Change ◽  
Learning Phase ◽  
Small Scale ◽  
Specific Gene ◽  
Expression Data ◽  
Validation Phase

Background Fatigue is a common side effect of cancer (CA) treatment. We used a novel analytical method to identify and validate a specific gene cluster that is predictive of fatigue risk in prostate cancer patients (PCP) treated with radiotherapy (RT). Methods A total of 44 PCP were categorized into high-fatigue (HF) and low-fatigue (LF) cohorts based on fatigue score change from baseline to RT completion. Fold-change differential and Fisher's linear discriminant analyses (LDA) from 27 subjects with gene expression data at baseline and RT completion generated a reduced base of most discriminatory genes (learning phase). A nearest-neighbor risk (k-NN) prediction model was developed based on small-scale prognostic signatures. The predictive model validity was tested in another 17 subjects using baseline gene expression data (validation phase). Result The model generated in the learning phase predicted HF classification at RT completion in the validation phase with 76.5% accuracy. Conclusion The results suggest that a novel analytical algorithm that incorporates fold-change differential analysis, LDA, and a k-NN may have applicability in predicting regimen-related toxicity in cancer patients with high reliability, if we take into account these results and the limited amount of data that we had at disposal. It is expected that the accuracy will be improved by increasing data sampling in the learning phase.

scholarly journals Investigating the Molecular Processes behind the Cell-Specific Toxicity Response to Titanium Dioxide Nanobelts

2021 ◽  
Vol 22 (17) ◽  
pp. 9432
Author(s):  
Laurent A. Winckers ◽  
Chris T. Evelo ◽  
Egon L. Willighagen ◽  
Martina Kutmon
Keyword(s):  
Gene Expression ◽  
Titanium Dioxide ◽  
Cell Lines ◽  
Gene Expression Data ◽  
Gene Networks ◽  
Biological Processes ◽  
Expression Data ◽  
Molecular Processes ◽  
High Exposure ◽  
Pathway Gene

Some engineered nanomaterials incite toxicological effects, but the underlying molecular processes are understudied. The varied physicochemical properties cause different initial molecular interactions, complicating toxicological predictions. Gene expression data allow us to study the responses of genes and biological processes. Overrepresentation analysis identifies enriched biological processes using the experimental data but prompts broad results instead of detailed toxicological processes. We demonstrate a targeted filtering approach to compare public gene expression data for low and high exposure on three cell lines to titanium dioxide nanobelts. Our workflow finds cell and concentration-specific changes in affected pathways linked to four Gene Ontology terms (apoptosis, inflammation, DNA damage, and oxidative stress) to select pathways with a clear toxicity focus. We saw more differentially expressed genes at higher exposure, but our analysis identifies clear differences between the cell lines in affected processes. Colorectal adenocarcinoma cells showed resilience to both concentrations. Small airway epithelial cells displayed a cytotoxic response to the high concentration, but not as strongly as monocytic-like cells. The pathway-gene networks highlighted the gene overlap between altered toxicity-related pathways. The automated workflow is flexible and can focus on other biological processes by selecting other GO terms.

Sign in / Sign up

Export Citation Format

Share Document