Skip to content

ADMET calculations from different research paper and a goal to make it memory efficent with ML implementations.

Notifications You must be signed in to change notification settings

santuchal/ADMET_calculations

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 

Repository files navigation

ADMET prediction using ML

This post is a step-by-step implementation of my approach to calculating the ADME(T) properties and it is meant for sharing, studying, and critiquing by fellow researchers who are new and interested in this topic. Most of the research paper that is implemented can be found in ref directory. 

NOTE

Note of the advanced machine learning module is upload in github and save as private repo.

  • marked is completed a long time ago.

Physicochemical Properties:

  • Molecular Weight
  • Volume
  • Density
  • Heavy Atoms
  • Aromatic Heavy Atoms
  • Fraction Csp3
  • Rotatable Bonds
  • Hetro Atoms
  • H-Bond Acceptors
  • H-Bond Donors
  • Ring Count
  • Aromatic Ring Count
  • Stereo Centers
  • Molar Refractivity
  • tPSA
  • LogP
  • pKa
  • LogS
  • LogD7.4

Medicinal Chemistry:


Absorption:

  • Caco-2 Permeability Using RF and rdkit and chemdesc descriptor
  • MDCK Permeability
  • HIA Using RF and MACCS Fingerprint
  • Pgp-substrate Using RF and SVM with ECFP4, ECFP6, MACCS
  • Pgp-inhibitor Using RF and SVM with ECFP4, ECFP6, MACCS
  • Oral Bioavailability
  • F20% Using RF and MACCS Fingerprint
  • F30% Using RF and MACCS Fingerprint

Distribution:

  • BBB Using SVM and ECFP6 with Morgan Fingerprint
  • PPB Using RF and rdkit and chemdesc descriptor
  • VD Using RF and rdkit and chemdesc descriptor

Metabolism:

  • CYP450 inhibitor/substrate
  • 1A2 Using RF and SVM with ECFP4, ECFP6, MorganFingerprint ( Different Accuracy, Precesion, FPR, Specifivity, ACC, F1 Score)
  • 2C9Using RF and SVM with ECFP4, ECFP6, MorganFingerprint ( Different Accuracy, Precesion, FPR, Specifivity, ACC, F1 Score)
  • 2C19Using RF and SVM with ECFP4, ECFP6, MorganFingerprint ( Different Accuracy, Precesion, FPR, Specifivity, ACC, F1 Score)
  • 2D6Using RF and SVM with ECFP4, ECFP6, MorganFingerprint ( Different Accuracy, Precesion, FPR, Specifivity, ACC, F1 Score)
  • 3A4Using RF and SVM with ECFP4, ECFP6, MorganFingerprint ( Different Accuracy, Precesion, FPR, Specifivity, ACC, F1 Score)

Excretion:

  • Half Life(T1/2) Using RF with descriptor(rdkit and chemdesc)
  • CL Using RF with descriptor(rdkit and chemdesc)

Toxicity:

  • Rodent Acute Toxicity
  • Human Acute Toxicity
  • Genotoxic Carcinogenicity Rule
  • Carcinogenicity
  • Skin Sensitization Rule
  • Aquatic Toxicity Rule
  • NonBiodegradable Rule
  • Sure ChEMBL Rule
  • FAF-Drugs4 Rule
  • Ames Mutagenicity
  • Hepatotoxicity Using RF and descriptor
  • Eye Corrosion
  • Eye Irritation
  • Respiratory Toxicity
  • Bioconcentration Factor
  • hERG inhibition/blockers Using Using RF and descriptor with MACCS
  • IGC50
  • LD50 Using RF 2D descriptor
  • AMES Using RF and MACCS
  • AR-LBD
  • SkinSen Using RF and MACCS
  • ER-LBD
  • Aromatase
  • PPAR-Y
  • DILI Using RF and MACCS
  • FDAMDO Using RF and ECFP4
  • p53
  • ARE
  • HSE
  • ATAD5

About

ADMET calculations from different research paper and a goal to make it memory efficent with ML implementations.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published