![ADME](https://storage.googleapis.com/polaris-public/icons/icons8-whale-96-ADME.png) 

## Background

The goal of accessing ADME properties is to understand how a potential drug candidate interacts with the human body, including absorption, distribution, metabolism, and excretion. This knowledge is crucial for evaluating efficacy, safety, and clinical potential, guiding drug development for optimal therapeutic outcomes. [Fang et al. 2023](https://doi.org/10.1021/acs.jcim.3c00160) has disclosed DMPK datasets collected over 20 months across six ADME in vitro endpoints, which are human and rat liver microsomal stability, MDR1-MDCK efflux ratio, solubility, and human and rat plasma protein binding. The dataset contains 885 to 3087 measures for the corresponding endpoints. 


## Benchmarking
**The goal** of this benchmark is to perform a multi task, which is to the best models for predicting six adme endpoints altogether. 

## Molecule data resource:
**Reference**: https://doi.org/10.1021/acs.jcim.3c00160

## Train/test split
To discover more potential hits which are similar to the discovered hits, a random splitting was applied. 


**Distribution of the train/test in the chemical space**
![image](https://storage.googleapis.com/polaris-public/benchmarks/figures/fang2023_ADME_public_v1_umap_random_split.png)

## Related links
The full curation and creation process is documented -> [notebook](https://github.com/polaris-hub/polaris-recipes/blob/main/02_MolProp/02_ADME_dataset.ipynb).