TCOF 3 :Repositioning of Chemical compounds From Different Classes as part of Virtual Screening

TCOF 3 :Repositioning of Chemical compounds From Different Classes as part of Virtual Screening

Under the Guidance of PI: Dr UCA JALEEL, Dr Bheemarao Ugarkar

(IISc Research Unit, Bangalore)

Yatindra Nath Yadav

3.4TCOF Fellow

(MSc Tech Bioinformatics, WBUT,KOLKATA)

Blog Url: yatindradotnet.wordpress.com

The aim of this project is to develop classes of anti MTb compounds and reposition them by screening pesticides which are found active against TB which we can further proceed with clinical trials.

Repositioning of Chemical compound database divided under three sub classes:-

1)Pesticides

2)Antimicrobial molecules

3)Phytomolecules

-> Me and My Group worked on Pesticides showing anti TB activity:•In search of Pesticide database we started with many search engine like Pubchem,PAN (Pesticide Action Network) pesticide database,Eu Pesticide Database & finally our search comes to an end with EPA (Environmental Protection Agency).

•In EPA we got some 654 pesticide molecules out of which we have structure and SDF file for 487 molecules remaining structure is drawn by (Ayisha safeeda) with the help of “MARVIN” and saved in SDF file format.

Data Scientific Authentication

For getting scientific background against 657 EPA registered Pesticide molecules we dropped couple of mails to following contacts:-

1) [email protected],2) [email protected],3) [email protected]

4) [email protected]

Next Slide will give an over view of the Project in the form of Flowchart to Explain the process.

AID 1332

Upload the sdf file

All compounds

sdf file

Generate descriptor file

Open the CSV file in Excel

Testing

Training File splitting

Remove the useless attributes

Select the actives and inactive compounds

Apply classifier

algorithms

Selection of best classifier

model

TP %, FP<20%, Accuracy >70%

Append the bioassay result corresponding

to the compounds

PubChem

PowerMV PowerMV

Excel

WEKA(machine learning)

Module – Work Flow

Current Stage of Project is Tuning of Model Generated by WEKA:

We are trying to Tune the Model (selecting best classifier)to the Most Stable state Applying the Cost Matrix on it .

We have generated the Results using different Classifiers like Naïve bayes and Random Forest We are trying to Tune the Model giving the Cost Matrix to it as shown in above excel sheet.

Next Stage is to Go for Screening and then We will proceed Further ….

References: 1) Schierz AC. Virtual screening of bioassay data. J Cheminform. 2009 Dec 22;1:21. doi: 10.1186/1758-2946-1-21. PubMed PMID: 20150999. 2) Periwal V etal., Predictive models for anti-tubercular molecules using machine learning on high-throughput biological screening datasets. BMC Res Notes. 2011 Nov18;4:504. doi: 10.1186/1756-0500-4-504. PubMed PMID: 22099929. 4) Enviornmental Protection Agency (EPA).

We are indebted to and earnestly acknowledge

Prof Dr Samir K Bramachari Dr TS BalganeshDr. U.C. Jaleel, PI (TOCF 3)Dr Bheemarao UgarkarIISc Research Unit, BangaloreOSDD open lab teamGroup Members (Swati shah,Ayisha Safeeda & Nufail)

Some Feed Back Needed Regarding Proposed Idea can we follow this

approach….????

Some feed backs needed on this Idea and software from Team & PI

SAR (Structural Activity Relationship) w.r.t field Points Concept:- Seeing the Ligand the way they experienced by protein receptor:-Taking 3D

Molecular Electrostatic Potential(Field points) viz:-Positive charge, negative charge,shape,hydrophocity.

Now the present of Field points suggest that this is an area of the ligand which will form the favorable interaction with protein receptor provided off course the protein receptor have complimentary features.

Cluster View

DISPARITY MATRIXS

Activity Miner

Thanks for your patience

OSDD fleets towards the ultimate TB drug