Search research articles

Search research articles

Related Experiment Videos

Prediction errors in learning drug response from gene expression data - influence of labeling, sample size, and

Immanuel Bayer¹, Philip Groth, Sebastian Schneckener

¹Aachen Institute for Advanced Study in Computational Engineering Science (AICES), RWTH Aachen University, Aachen, Germany.

|July 30, 2013

Summary

This summary is machine-generated.

Related Concept Videos

Pharmacogenomics: Identification of New Drug Targets

Pharmacogenomics: Identification of New Drug Targets

Advances in genomics have profoundly influenced drug discovery by increasing both the speed and accuracy of pharmaceutical development. Pharmacogenomics, which examines how genetic variation influences drug response, facilitates the identification of novel therapeutic targets and enables patient stratification for personalized treatment. These strategies contribute to improved drug efficacy, minimized adverse effects, and more efficient clinical trial design.Mapping genetic differences...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Prediction of human pharmacokinetics from chemical structure: combining mechanistic modeling with machine learning.

Journal of pharmaceutical sciences·2024

Same author

A deep neural network: mechanistic hybrid model to predict pharmacokinetics in rat.

Journal of computer-aided molecular design·2024

Same author

Emodepside targets SLO-1 channels of Onchocerca ochengi and induces broad anthelmintic effects in a bovine model of onchocerciasis.

PLoS pathogens·2021

Same author

A workflow to build PBTK models for novel species.

Archives of toxicology·2020

Same author

Bayer's in silico ADMET platform: a journey of machine learning over the past two decades.

Drug discovery today·2020

Same author

Prediction of Oral Bioavailability in Rats: Transferring Insights from in Vitro Correlations to (Deep) Machine Learning Models Using in Silico Model Outputs and Chemical Structure Parameters.

Journal of chemical information and modeling·2019

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Predicting cancer cell sensitivity to drugs depends heavily on data quality and sample size. Focusing on robust sample collection and high-quality data is more impactful than algorithm selection for accurate drug response modeling.

Area of Science:

Computational biology
Pharmacogenomics
Machine learning in drug discovery

Background:

Model-based prediction for drug response relies on numerous factors, including data collection, prediction endpoints, and algorithm parameters.
Predicting cancer cell line sensitivity to various compounds (e.g., IC50 values) is crucial for personalized medicine.

Purpose of the Study:

To investigate the impact of different choices in model-based prediction, specifically focusing on sample collection, algorithms, and prediction endpoints for drug response.
To compare the performance of various machine learning algorithms and labeling strategies across different sample collections.

Main Methods:

Utilized three independent sample collections to predict drug sensitivity (IC50) in cancer cell lines.
Applied multiple machine learning algorithms and compared various prediction endpoints and labeling strategies.

Related Experiment Videos

Evaluated model performance against identically generated null models across diverse compound and cell line combinations.

Main Results:

The predictability of treatment effects varied significantly among compounds, with some showing high predictability and others low.
Sample collection choice and sample size were identified as major factors in reducing prediction error.
No single machine learning algorithm consistently outperformed others, and no significant difference was found between regression and classification (two- or three-class) predictors.

Conclusions:

Efforts in drug response modeling should prioritize enhancing sample collection and ensuring high data quality over extensive method or algorithm adjustment.
Robust data acquisition strategies are paramount for improving the accuracy and reliability of predictive models in cancer therapeutics.