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

Motivation: Many high-performance DTA models have been proposed, but they aremostly black-box and thus lack human interpretability. Explainable AI (XAI) canmake DTA models more trustworthy, and can also enable scientists to distillbiological knowledge from the models. Counterfactual explanation is one popularapproach to explaining the behaviour of a deep neural network, which works bysystematically answering the question "How would the model output change if theinputs were changed in this way?". Most counterfactual explanation methods onlyoperate on single input data. It remains an open problem how to extendcounterfactual-based XAI methods to DTA models, which have two inputs, one fordrug and one for target, that also happen to be discrete in nature.  Methods: We propose a multi-agent reinforcement learning framework,Multi-Agent Counterfactual Drug target binding Affinity (MACDA), to generatecounterfactual explanations for the drug-protein complex. Our proposedframework provides human-interpretable counterfactual instances whileoptimizing both the input drug and target for counterfactual generation at thesame time.  Results: We benchmark the proposed MACDA framework using the Davis datasetand find that our framework produces more parsimonious explanations with noloss in explanation validity, as measured by encoding similarity and QED. Wethen present a case study involving ABL1 and Nilotinib to demonstrate how MACDAcan explain the behaviour of a DTA model in the underlying substructureinteraction between inputs in its prediction, revealing mechanisms that alignwith prior domain knowledge.

Counterfactual Explanation with Multi-Agent Reinforcement Learning for Drug Target Prediction