English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Patient-derived xenografts (PDXs) are an appealing platform for preclinicaldrug studies because the in vivo environment of PDXs helps preserve tumorheterogeneity and usually better mimics drug response of patients with cancercompared to CCLs. We investigate multimodal neural network (MM-Net) and dataaugmentation for drug response prediction in PDXs. The MM-Net learns to predictresponse using drug descriptors, gene expressions (GE), and histologywhole-slide images (WSIs) where the multi-modality refers to the tumorfeatures. We explore whether the integration of WSIs with GE improvespredictions as compared with models that use GE alone. We use two methods toaddress the limited number of response values: 1) homogenize drugrepresentations which allows to combine single-drug and drug-pairs treatmentsinto a single dataset, 2) augment drug-pair samples by switching the order ofdrug features which doubles the sample size of all drug-pair samples. Thesemethods enable us to combine single-drug and drug-pair treatments, allowing usto train multimodal and unimodal neural networks (NNs) without changingarchitectures or the dataset. Prediction performance of three unimodal NNswhich use GE are compared to assess the contribution of data augmentationmethods. NN that uses the full dataset which includes the original and theaugmented drug-pair treatments as well as single-drug treatments significantlyoutperforms NNs that ignore either the augmented drug-pairs or the single-drugtreatments. In assessing the contribution of multimodal learning based on theMCC metric, MM-Net statistically significantly outperforms all the baselines.Our results show that data augmentation and integration of histology imageswith GE can improve prediction performance of drug response in PDXs.

Data augmentation and multimodal learning for predicting drug response in patient-derived xenografts from gene expressions and histology images