Model: eos60li - Aqueous Solubility Prediction

Overview

This repository contains the model validation task for the Outreachy Summer Internship 2024 contributions to Ersilia.

Model Information

• Input: Compound
• Input Shape: Single
• Task: git addRegression
• Output: Experimental value
• Output Type: Float
• Output Shape: Single
• Interpretation: Predicted LogS (log of the solubility)

Steps to Reproduce

1. Install Ersilia Model Hub

Install Ersilia Model Hub on your system here.

2. Download Model and Run

Run the following commands on your terminal to download and run the model on your system.

ersilia -v fetch eos6oli
ersilia -v serve eos6oli
ersilia api run -i data/input.csv -o data/output.csv

Dataset

The dataset used on the model for checking for model bias was downloaded from here. The dataset was from the research conducted by John S. Delaney on ESOL:  Estimating Aqueous Solubility Directly from Molecular Structure.

References

• Ersilia Model Hub
• ESOL:  Estimating Aqueous Solubility Directly from Molecular Structure
• SolTraNet