scDrug+: predicting drug-responses using single-cell transcriptomics and molecular structure - 23/07/24

Doi : 10.1016/j.biopha.2024.117070

Yih-Yun Sun ^a,^b, Chiao-Yu Hsieh ^b, Jian-Hung Wen ^b,^c, Tzu-Yang Tseng ^a,^d, Jia-Hsin Huang ^b, Yen-Jen Oyang ^a, Hsuan-Cheng Huang ^c,^⁎ , Hsueh-Fen Juan ^a,^b,^d,^e,^f,^⁎
^a Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taiwan
^b Taiwan AI Labs, Taipei 10351, Taiwan
^c Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei 11221, Taiwan
^d Department of Life Science, National Taiwan University, Taipei 106, Taiwan
^e Center for Computational and Systems Biology, National Taiwan University, Taipei 106, Taiwan
^f Center for Advanced Computing and Imaging in Biomedicine, National Taiwan University, Taipei 106, Taiwan

^⁎Corresponding author.^⁎⁎Corresponding author at: Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taiwan.Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan UniversityTaiwan

Abstract

Predicting drug responses based on individual transcriptomic profiles holds promise for refining prognosis and advancing precision medicine. Although many studies have endeavored to predict the responses of known drugs to novel transcriptomic profiles, research into predicting responses for newly discovered drugs remains sparse. In this study, we introduce scDrug+, a comprehensive pipeline that seamlessly integrates single-cell analysis with drug-response prediction. Importantly, scDrug+ is equipped to predict the response of new drugs by analyzing their molecular structures. The open-source tool is available as a Docker container, ensuring ease of deployment and reproducibility. It can be accessed at scDrugplus.

Il testo completo di questo articolo è disponibile in PDF.

Highlights

•	scDrug+ integrates single-cell analysis with drug-response prediction.
•	scDrug+ predicts responses of novel drugs on cell subpopulations.
•	Matrix factorization and SVM with molecular fingerprints show superior performance.

Il testo completo di questo articolo è disponibile in PDF.

Keywords : Drug-responses, Single-cell transcriptomics, Machine learning, Precision medicine

Mappa

Background

Results

Data preparation, model selection, and performance evaluation

In-depth analysis of molecular similarity, data integrity, and their impacts on drug response model performance

Impact of missing data on model performance and comparative analysis in drug response predictions

Comparative assessment of the combined model and precily on drug response prediction from GDSC and PRISM datasets

Discussion

Conclusions

Methods

Data integration and preprocessing: harnessing the PRISM, GDSC, and CCLE datasets for drug response predictions

Navigating molecular representations: from graph isomorphism to drug prediction models

From SMILES strings to neural embeddings: expanding the horizons of molecular representation

Deciphering molecular similarity and gene expression profiling: advanced methods for drug discovery

Harnessing molecular representations and machine learning for advanced drug response predictions from RDKit to GNNs

Enhancing drug response predictions: integrating molecular embeddings and modifying CaDRReS matrix factorization approaches

Comparative evaluation methods for drug response: cell-wise vs. drug-wise metrics

Funding

CRediT authorship contribution statement

Esportazione

Vol 177

Articolo 117070- agosto 2024 Ritorno al numero

Articolo precedente

Tanshinone I improves TNBC chemosensitivity by suppressing late-phase autophagy through AKT/p38 MAPK signaling pathway
Xinglinzi Tang, Jiaqian Gong, Linlin Ren, Zhiyu Wang, Bowen Yang, Wenzhu Wang, Neng Wang

| Articolo seguente

Apigenin improves testosterone synthesis by regulating endoplasmic reticulum stress
Liling Wu, Jiayuan Qu, Li Mou, Changjiang Liu

Benvenuto su EM|consulte, il riferimento dei professionisti della salute.
L'accesso al testo integrale di questo articolo richiede un abbonamento.

Già abbonato a @@106933@@ rivista ?

connettersi o creare un account

scDrug+: predicting drug-responses using single-cell transcriptomics and molecular structure - 23/07/24

Abstract

Highlights

Mappa

Citazioni Export

File

Contenuto

Il mio account

Aide & support

Dichiarazione CNIL