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

Borisov, Nicolas; Tkachev, Victor; Suntsova, Maria; Kovalchuk, Olga; Zhavoronkov, Alex; Muchnik, Ilya; Buzdin, Anton

doi:10.1080/15384101.2017.1417706

Cited by 55 publications

(52 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach was initially tested for the SVM method [8,[33][34][35], and in this study, we for the first time applied it to supplement other six popular ML techniques. We used twenty-one clinically annotated gene expression datasets totally, including 1778 patient samples with known clinical treatment responses.…”

Section: Discussionmentioning

confidence: 99%

“…This approach implies that the ML model is trained on a bigger, similar, but quite different, dataset, and then applied to a smaller (validation) dataset. The FloWPS technique has been already tested for transfer learning, and gene expression profiles of cell cultures treated with chemotherapeutic drugs served as training datasets [33][34][35]. Another possibility is to aggregate different smaller datasets into bigger ones.…”

Section: Discussionmentioning

confidence: 99%

“…We first introduced [33][34][35] flexible data trimming as a preprocessing tool for transferring to real patients the gene expression data obtained for cell cultures treated with anti-cancer drugs.…”

Section: Principles Of Flexible Data Trimmingmentioning

confidence: 99%

“…For generating statistically significant predictions, extensive reduction of a pool of features under consideration is needed to make their number comparable with the number of individual samples available [10,[29][30][31]. To leverage the performance of ML in biomedicine, we recently developed an approach called flexible data trimming (Data trimming (DT) is the process of removing or excluding extreme values, or outliers, from a dataset [32]) [8,29,[33][34][35]. This approach is heuristic and based on a common geometrical sense ( Figure 1).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Flexible Data Trimming Improves Performance of Global Machine Learning Methods in Omics-Based Personalized Oncology

Tkachev

Sorokin

Borisov

et al. 2020

IJMS

Self Cite

View full text Add to dashboard Cite

(1) Background: Machine learning (ML) methods are rarely used for an omics-based prescription of cancer drugs, due to shortage of case histories with clinical outcome supplemented by high-throughput molecular data. This causes overtraining and high vulnerability of most ML methods. Recently, we proposed a hybrid global-local approach to ML termed floating window projective separator (FloWPS) that avoids extrapolation in the feature space. Its core property is data trimming, i.e., sample-specific removal of irrelevant features. (2) Methods: Here, we applied FloWPS to seven popular ML methods, including linear SVM, k nearest neighbors (kNN), random forest (RF), Tikhonov (ridge) regression (RR), binomial naïve Bayes (BNB), adaptive boosting (ADA) and multi-layer perceptron (MLP). (3) Results: We performed computational experiments for 21 high throughput gene expression datasets (41–235 samples per dataset) totally representing 1778 cancer patients with known responses on chemotherapy treatments. FloWPS essentially improved the classifier quality for all global ML methods (SVM, RF, BNB, ADA, MLP), where the area under the receiver-operator curve (ROC AUC) for the treatment response classifiers increased from 0.61–0.88 range to 0.70–0.94. We tested FloWPS-empowered methods for overtraining by interrogating the importance of different features for different ML methods in the same model datasets. (4) Conclusions: We showed that FloWPS increases the correlation of feature importance between the different ML methods, which indicates its robustness to overtraining. For all the datasets tested, the best performance of FloWPS data trimming was observed for the BNB method, which can be valuable for further building of ML classifiers in personalized oncology.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Principles Of Flexible Data Trimmingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Flexible Data Trimming Improves Performance of Global Machine Learning Methods in Omics-Based Personalized Oncology

Tkachev

Sorokin

Borisov

et al. 2020

IJMS

Self Cite

View full text Add to dashboard Cite

show abstract

“…GSE27448 [19,20] 5 10 5251 4701 GSE52519 [21] 3 9 751 742 GSE61615 [22] 2 2 842 736 GSE76211 [23,24] 3 3 770 658 GSE100926 [25] 3 3 223 194 TCGA-BLCA [13] 19 406 2873 2537…”

Section: Number Of Degs After Funrich Mappingmentioning

confidence: 99%

Bioinformatics Analysis Identifying Key Biomarkers in Bladder Cancer

Zhang

Berndt‐Paetz

Neuhaus

2020

Data

View full text Add to dashboard Cite

Our goal was to find new diagnostic and prognostic biomarkers in bladder cancer (BCa), and to predict molecular mechanisms and processes involved in BCa development and progression. Notably, the data collection is an inevitable step and time-consuming work. Furthermore, identification of the complementary results and considerable literature retrieval were requested. Here, we provide detailed information of the used datasets, the study design, and on data mining. We analyzed differentially expressed genes (DEGs) in the different datasets and the most important hub genes were retrieved. We report on the meta-data information of the population, such as gender, race, tumor stage, and the expression levels of the hub genes. We include comprehensive information about the gene ontology (GO) enrichment analyses and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses. We also retrieved information about the up-and down-regulation of genes. All in all, the presented datasets can be used to evaluate potential biomarkers and to predict the performance of different preclinical biomarkers in BCa. Dataset:The following are available online at

show abstract

Shambhala‐2: A Protocol for Uniformly Shaped Harmonization of Gene Expression Profiles of Various Formats

et al. 2022

Self Cite

View full text Add to dashboard Cite

Uniformly shaped harmonization of gene expression profiles is central for the simultaneous comparison of multiple gene expression datasets. It is expected to operate with the gene expression data obtained using various experimental methods and equipment, and to return harmonized profiles in a uniform shape. Such uniformly shaped expression profiles from different initial datasets can be further compared directly. However, current harmonization techniques have strong limitations that prevent their broad use for bioinformatic applications. They can either operate with only up to two datasets/platforms or return data in a dynamic format that will be different for every comparison under analysis. This also does not allow for adding new data to the previously harmonized dataset(s), which complicates the analysis and increases calculation costs. We propose here a new method termed Shambhala‐2 that can transform multi‐platform expression data into a universal format that is identical for all harmonizations made using this technique. Shambhala‐2 is based on sample‐by‐sample cubic conversion of the initial expression dataset into a preselected shape of the reference definitive dataset. Using 8390 samples of 12 healthy human tissue types and 4086 samples of colorectal, kidney, and lung cancer tissues, we verified Shambhala‐2's capacity in restoring tissue‐specific expression patterns for seven microarray and three RNA sequencing platforms. Shambhala‐2 performed well for all tested combinations of RNAseq and microarray profiles, and retained gene‐expression ranks, as evidenced by high correlations between different single‐ or aggregated gene expression metrics in pre‐ and post‐Shambhalized samples, including preserving cancer‐specific gene expression and pathway activation features. © 2022 Wiley Periodicals LLC. Basic Protocol: Shambhala‐2 harmonizer Alternate Protocol 1: Linear Shambhala/Shambhala‐1 Alternate Protocol 2: Alternative (flexible‐format and uniformly shaped) normalization methods Support Protocol 1: Watermelon multisection (WM) Support Protocol 2: Calculation of cancer‐to‐normal log‐fold‐change (LFC) and pathway activation level (PAL)

show abstract

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

Cited by 55 publications

References 24 publications

Flexible Data Trimming Improves Performance of Global Machine Learning Methods in Omics-Based Personalized Oncology

Flexible Data Trimming Improves Performance of Global Machine Learning Methods in Omics-Based Personalized Oncology

Bioinformatics Analysis Identifying Key Biomarkers in Bladder Cancer

Shambhala‐2: A Protocol for Uniformly Shaped Harmonization of Gene Expression Profiles of Various Formats

Contact Info

Product

Resources

About