Classifying Breast Cancer Subtypes Using Deep Neural Networks Based on Multi-Omics Data

Lin, Yuqi; Zhang, Wen; Cao, Huanshen; Li, Gaoyang; Du, Wei

doi:10.3390/genes11080888

Cited by 67 publications

(52 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among the deep learning models developed for multi-omics integration, we can present MOLI [13] , [127] , which retrieved DL-based features using subnetworks on each omics dataset and concatenate the obtained deep features. Then, a final neural network is used on the concatenated deep features for prediction of drug activity.…”

Section: Main Integration Strategiesmentioning

confidence: 99%

Integration strategies of multi-omics data for machine learning analysis

Picard

Scott‐Boyer

Bodein

et al. 2021

Computational and Structural Biotechnology Journal

273

185

View full text Add to dashboard Cite

Graphical abstract Schematic representation of the main strategies for multi-omics datasets integration. A) Early integration concatenates all omics datasets into a single matrix on which machine learning model can be applied. B) Mixed integration first independently transforms or maps each omics block into a new representation before combining them for downstream analysis. C) Intermediate integration simultaneously transforms the original datasets into common and omics-specific representations. D) Late integration analyses each omics separately and combines their final predictions. E) Hierarchical integration bases the integration of datasets on prior regulatory relationships between omics layers.

show abstract

Section: Main Integration Strategiesmentioning

confidence: 99%

Integration strategies of multi-omics data for machine learning analysis

Picard

Scott‐Boyer

Bodein

et al. 2021

Computational and Structural Biotechnology Journal

273

185

View full text Add to dashboard Cite

show abstract

“…The determination of the molecular subtype is commonly performed by immunohistochemistry and/or genetic analyses and its classification is related to the positivity or negativity for estrogen and progesterone receptors (ER and PR), as well as for eventual (over)expression of the oncogene human epidermal growth factor receptor 2 (HER-2). The main molecular subtypes are: (a) luminal (ER and PR positive); (b) HER-2 enriched (ER, PR negative, and HER-2 overexpression), and (c) triple-negative breast cancer (TNBC) (ER, PR, and HER-2 negative) [ 4 , 5 , 6 ]. For luminal and HER-2 subtypes, there are effective therapeutic drugs [ 7 ], such as the well-established ER antagonist tamoxifen for hormone-positive tumors [ 8 ] and the antibody trastuzumab, to HER2 subtype [ 9 ].…”

Section: Introductionmentioning

confidence: 99%

Cytotoxicity of Seaweed Compounds, Alone or Combined to Reference Drugs, against Breast Cell Lines Cultured in 2D and 3D

Malhão

Ramos

Macedo

et al. 2021

Toxics

View full text Add to dashboard Cite

Seaweed bioactive compounds have shown anticancer activities in in vitro and in vivo studies. However, tests remain limited, with conflicting results, and effects in combination with anticancer drugs are even scarcer. Here, the cytotoxic effects of five seaweed compounds (astaxanthin, fucoidan, fucosterol, laminarin, and phloroglucinol) were tested alone and in combination with anticancer drugs (cisplatin—Cis; and doxorubicin—Dox), in breast cell lines (three breast cancer (BC) subtypes and one non-tumoral). The combinations revealed situations where seaweed compounds presented potentiation or inhibition of the drugs’ cytotoxicity, without a specific pattern, varying according to the cell line, concentration used for the combination, and drug. Fucosterol was the most promising compound, since: (i) it alone had the highest cytotoxicity at low concentrations against the BC lines without affecting the non-tumoral line; and (ii) in combination (at non-cytotoxic concentration), it potentiated Dox cytotoxicity in the triple-negative BC cell line. Using a comparative approach, monolayer versus 3D cultures, further investigation assessed effects on cell viability and proliferation, morphology, and immunocytochemistry targets. The cytotoxic and antiproliferative effects in monolayer were not observed in 3D, corroborating that cells in 3D culture are more resistant to treatments, and reinforcing the use of more complex models for drug screening and a multi-approach that should include histological and ICC analysis.

show abstract

“…In this study, although Super.FELT was applied for the drug response prediction, it can be further applied for other biomedical tasks using multi-omics datasets. Recently, disease progress prediction, such as survival and recurrence, and cancer subtype classification, has been performed using multi-omics datasets [52][53][54][55]. In those studies, AE, a chi-squared test, and a feedforward network have been used to represent features in omics data.…”

Section: Discussionmentioning

confidence: 99%

Super.FELT: supervised feature extraction learning using triplet loss for drug response prediction with multi-omics data

2021

View full text Add to dashboard Cite

Background Predicting the drug response of a patient is important for precision oncology. In recent studies, multi-omics data have been used to improve the prediction accuracy of drug response. Although multi-omics data are good resources for drug response prediction, the large dimension of data tends to hinder performance improvement. In this study, we aimed to develop a new method, which can effectively reduce the large dimension of data, based on the supervised deep learning model for predicting drug response. Results We proposed a novel method called Supervised Feature Extraction Learning using Triplet loss (Super.FELT) for drug response prediction. Super.FELT consists of three stages, namely, feature selection, feature encoding using a supervised method, and binary classification of drug response (sensitive or resistant). We used multi-omics data including mutation, copy number aberration, and gene expression, and these were obtained from cell lines [Genomics of Drug Sensitivity in Cancer (GDSC), Cancer Cell Line Encyclopedia (CCLE), and Cancer Therapeutics Response Portal (CTRP)], patient-derived tumor xenografts (PDX), and The Cancer Genome Atlas (TCGA). GDSC was used for training and cross-validation tests, and CCLE, CTRP, PDX, and TCGA were used for external validation. We performed ablation studies for the three stages and verified that the use of multi-omics data guarantees better performance of drug response prediction. Our results verified that Super.FELT outperformed the other methods at external validation on PDX and TCGA and was good at cross-validation on GDSC and external validation on CCLE and CTRP. In addition, through our experiments, we confirmed that using multi-omics data is useful for external non-cell line data. Conclusion By separating the three stages, Super.FELT achieved better performance than the other methods. Through our results, we found that it is important to train encoders and a classifier independently, especially for external test on PDX and TCGA. Moreover, although gene expression is the most powerful data on cell line data, multi-omics promises better performance for external validation on non-cell line data than gene expression data. Source codes of Super.FELT are available at https://github.com/DMCB-GIST/Super.FELT.

show abstract

Classifying Breast Cancer Subtypes Using Deep Neural Networks Based on Multi-Omics Data

Cited by 67 publications

References 42 publications

Integration strategies of multi-omics data for machine learning analysis

Integration strategies of multi-omics data for machine learning analysis

Cytotoxicity of Seaweed Compounds, Alone or Combined to Reference Drugs, against Breast Cell Lines Cultured in 2D and 3D

Super.FELT: supervised feature extraction learning using triplet loss for drug response prediction with multi-omics data

Contact Info

Product

Resources

About