Molecule Property Optimizer

AstraZeneca - MAI |

Molecule Property Optimizer

Drug Design

Run
About
API Example

Step 1: Upload your data

Upload Source Molecules

Drag your file(s) or upload

Your file can be in the following formats:csv

Your input data should be .csv or .txt and comma-separated. And, your input data must have the following column headings: 1. "Source_Mol" (Your source molecule in SMILES format), 2. "Source_Mol_LogD" (Your source molecule’s logD e.g. 2.52340477180304), 3. "Target_Mol_LogD" (The LogD you want your molecule to have), 4. "Source_Mol_Solubility" (Your source molecule’s solubility e.g. 1.86827115391515), 5. "Target_Mol_Solubility" (The solubility you want your molecule to have), 6. "Source_Mol_Clint" (Your source molecule’s clearance e.g. 2.075152704), 7. "Target_Mol_Clint" (The clearance you want your molecule to have).

Don’t have a file?

Use our demo data to run

Use Demo Data

View example data

Step 2: Set Parameters

Number of molecules to generate

Draw generated molecules

Step 3: Complete run profile

Job name - Optional

Generates new molecules with desirable properties given a starting molecule, essentially replicating a chemist's intuition to optimize molecular properties, accelerating drug discovery

Example use case: molecular optimization for finding drug candidates

Limitations: works with 3 ADMET properties: logD, solubility and clearance

Technology: Message passing neural network

Metrics: As reported by He et al.

Citation:

He J, You H, Sandström E, Nittinger E, Bjerrum EJ, Tyrchan C, Czechtizky W, Engkvist O. Molecular optimization by capturing chemist's intuition using deep neural networks. J Cheminform. 2021 Mar 20;13(1):26. doi: 10.1186/s13321-021-00497-0. PMID: 33743817; PMCID: PMC7980633.

Released:
Aug-30-2022

Previous Job Parameters

Your previous job parameters will show up here
so you can keep track of your jobs

Results

Parameters