date: 2022-08-22T03:46:16Z
pdf:unmappedUnicodeCharsPerPage: 0
pdf:PDFVersion: 1.7
pdf:docinfo:title: Pre-Training on In Vitro and Fine-Tuning on Patient-Derived Data Improves Deep Neural Networks for Anti-Cancer Drug-Sensitivity Prediction
xmp:CreatorTool: LaTeX with hyperref
Keywords: deep neural networks; drug-sensitivity prediction; anti-cancer drugs
access_permission:modify_annotations: true
access_permission:can_print_degraded: true
subject: Large-scale databases that report the inhibitory capacities of many combinations of candidate drug compounds and cultivated cancer cell lines have driven the development of preclinical drug-sensitivity models based on machine learning. However, cultivated cell lines have devolved from human cancer cells over years or even decades under selective pressure in culture conditions. Moreover, models that have been trained on in vitro data cannot account for interactions with other types of cells. Drug-response data that are based on patient-derived cell cultures, xenografts, and organoids, on the other hand, are not available in the quantities that are needed to train high-capacity machine-learning models. We found that pre-training deep neural network models of drug sensitivity on in vitro drug-sensitivity databases before fine-tuning the model parameters on patient-derived data improves the models' accuracy and improves the biological plausibility of the features, compared to training only on patient-derived data. From our experiments, we can conclude that pre-trained models outperform models that have been trained on the target domains in the vast majority of cases.
dc:creator: Paul Prasse, Pascal Iversen, Matthias Lienhard, Kristina Thedinga, Ralf Herwig and Tobias Scheffer
dcterms:created: 2022-08-22T03:41:16Z
Last-Modified: 2022-08-22T03:46:16Z
dcterms:modified: 2022-08-22T03:46:16Z
dc:format: application/pdf; version=1.7
title: Pre-Training on In Vitro and Fine-Tuning on Patient-Derived Data Improves Deep Neural Networks for Anti-Cancer Drug-Sensitivity Prediction
Last-Save-Date: 2022-08-22T03:46:16Z
pdf:docinfo:creator_tool: LaTeX with hyperref
access_permission:fill_in_form: true
pdf:docinfo:keywords: deep neural networks; drug-sensitivity prediction; anti-cancer drugs
pdf:docinfo:modified: 2022-08-22T03:46:16Z
meta:save-date: 2022-08-22T03:46:16Z
pdf:encrypted: false
dc:title: Pre-Training on In Vitro and Fine-Tuning on Patient-Derived Data Improves Deep Neural Networks for Anti-Cancer Drug-Sensitivity Prediction
modified: 2022-08-22T03:46:16Z
cp:subject: Large-scale databases that report the inhibitory capacities of many combinations of candidate drug compounds and cultivated cancer cell lines have driven the development of preclinical drug-sensitivity models based on machine learning. However, cultivated cell lines have devolved from human cancer cells over years or even decades under selective pressure in culture conditions. Moreover, models that have been trained on in vitro data cannot account for interactions with other types of cells. Drug-response data that are based on patient-derived cell cultures, xenografts, and organoids, on the other hand, are not available in the quantities that are needed to train high-capacity machine-learning models. We found that pre-training deep neural network models of drug sensitivity on in vitro drug-sensitivity databases before fine-tuning the model parameters on patient-derived data improves the models' accuracy and improves the biological plausibility of the features, compared to training only on patient-derived data. From our experiments, we can conclude that pre-trained models outperform models that have been trained on the target domains in the vast majority of cases.
pdf:docinfo:subject: Large-scale databases that report the inhibitory capacities of many combinations of candidate drug compounds and cultivated cancer cell lines have driven the development of preclinical drug-sensitivity models based on machine learning. However, cultivated cell lines have devolved from human cancer cells over years or even decades under selective pressure in culture conditions. Moreover, models that have been trained on in vitro data cannot account for interactions with other types of cells. Drug-response data that are based on patient-derived cell cultures, xenografts, and organoids, on the other hand, are not available in the quantities that are needed to train high-capacity machine-learning models. We found that pre-training deep neural network models of drug sensitivity on in vitro drug-sensitivity databases before fine-tuning the model parameters on patient-derived data improves the models' accuracy and improves the biological plausibility of the features, compared to training only on patient-derived data. From our experiments, we can conclude that pre-trained models outperform models that have been trained on the target domains in the vast majority of cases.
Content-Type: application/pdf
pdf:docinfo:creator: Paul Prasse, Pascal Iversen, Matthias Lienhard, Kristina Thedinga, Ralf Herwig and Tobias Scheffer
X-Parsed-By: org.apache.tika.parser.DefaultParser
creator: Paul Prasse, Pascal Iversen, Matthias Lienhard, Kristina Thedinga, Ralf Herwig and Tobias Scheffer
meta:author: Paul Prasse, Pascal Iversen, Matthias Lienhard, Kristina Thedinga, Ralf Herwig and Tobias Scheffer
dc:subject: deep neural networks; drug-sensitivity prediction; anti-cancer drugs
meta:creation-date: 2022-08-22T03:41:16Z
created: 2022-08-22T03:41:16Z
access_permission:extract_for_accessibility: true
access_permission:assemble_document: true
xmpTPg:NPages: 14
Creation-Date: 2022-08-22T03:41:16Z
pdf:charsPerPage: 3844
access_permission:extract_content: true
access_permission:can_print: true
meta:keyword: deep neural networks; drug-sensitivity prediction; anti-cancer drugs
Author: Paul Prasse, Pascal Iversen, Matthias Lienhard, Kristina Thedinga, Ralf Herwig and Tobias Scheffer
producer: pdfTeX-1.40.21
access_permission:can_modify: true
pdf:docinfo:producer: pdfTeX-1.40.21
pdf:docinfo:created: 2022-08-22T03:41:16Z