Certara, Inc. · NASDAQ:CERT
Certara — Technical Presentation (2014)
39 slides from Certara's technical presentation, 2014
Slide 1: Reaction Driven Molecular Invention - Generating Synthetically Feasible Design Ideas
Reaction Driven Molecular Invention - Generating Synthetically Feasible Design Ideas Brian B. Masek, Ph.D.
Slide 2: Today's Presenter
Today’s Presenter • Brian Masek, Ph.D. – Product Manager and Lead Scientist – Development and validation of novel computer-assisted drug discovery methods – Current focus on molecular invention and multi-criteria drug design – Prior to Certara, Director for Scientific R&D at Optive Research and posi
Slide 3: The Problem
The Problem Compounds exist that have some desirable properties to become a drug but… The known compounds are not suitable due to: – Intellectual Property – Physical Properties – DMPK Properties – Toxicological Properties – Each with its own SAR!! 3 © Copyright 2014 Certara, L.P. All rights reserved
Slide 4: The Design Challenge
The Design Challenge What compound should I make next? How to generate ideas that will satisfy ALL the criteria for a successful drug candidate? Must have design ideas that can be pursued in the lab 4 © Copyright 2014 Certara, L.P. All rights reserved.
Slide 5: Molecular Evolution (aka de novo design)
Molecular Evolution (aka de novo design) Potential access to a universe of chemistry But not all areas of the universe are interesting Challenge has been to focus on medicinally relevant structures Quality of the structures is VERY important Synthetic accessibility is VERY important 5 © Copyright 20
Slide 6: Multi-Criteria Design - Neurocrine
Multi-Criteria Design - Neurocrine Invention of Selective Norephinephrine Re-uptake Inhibitors O O H O N O O O N Training Set O O H O O H N O N H H ES 2131020 WO 04018441 WO 04018441 CF3 N H O Atomoxetine Reboxetine Muse Invent “Engine” O NH NH NH H N O MDL 27777A ES 2131020 DE 2627227 HN US 6759435
Slide 7: An Example: Estradiol
An Example: Estradiol I know about Estradiol I want to generate ideas for novel estrogenic compounds © Copyright 2014 Certara, L.P. All rights reserved. 7
Slide 8: © Copyright 2014 Certara, L.P. All rights reserved.
© Copyright 2014 Certara, L.P. All rights reserved. 8
Slide 9: © Copyright 2014 Certara, L.P. All rights reserved.
© Copyright 2014 Certara, L.P. All rights reserved. 9
Slide 10: The Reaction Driven De Novo Design Process
The Reaction Driven De Novo Design Process 10 © Copyright 2014 Certara, L.P. All rights reserved.
Slide 11: Molecular Evolution (aka de novo design)
Molecular Evolution (aka de novo design) Potential access to a universe of chemistry But not all areas of the universe are interesting Challenge has been to focus on medicinally relevant structures Quality of the structures is VERY important Synthetic accessibility is VERY important 11 © Copyright 2
Slide 12: Previous Work to Meld Synthesis with Design
Previous Work to Meld Synthesis with Design Pioneering work by: – Cramer et. al. Exhaustive enumeration of a synthetic chemistry space Search piecewise with Topomer Similarity AllChem: generating and searching 1020 synthetically accessible structures J. Comp-Aided Mol. Des. 2007, 21, 341-350. – G. S
Slide 13: Overview of Molecular Evolution Process
Overview of Molecular Evolution Process Seed Structures Invent Population Population of Parents Of Children (With Scores) Repeat for Number of Generations Result Decouple “Invention from Structures “Scoring” Select Population Of Parents Score – Scoring can be literally any & Children CADD design cri
Slide 14: Rxn Driven De Novo Design Process
Rxn Driven De Novo Design Process “Dream” Mode” • Rich set of virtual chemistry “Synthetic Chemistry” Mode • Reactions & Reactants Starting Materials Invent Population Population of Parents Of Children (With Scores) Repeat for Number of Generations Result Structures Select Population Of Parents Scor
Slide 15: Synthetic Chemistry Operators
Synthetic Chemistry Operators All Operators except for crossover have a corresponding “undo” operator Operators address either large scale changes in synthesis or small ones Large Scale Synthetic Operators Probability Add “M” synthetic steps (13.61%) Undo all steps (18.14%) Pick a new starting mater
Slide 16: The Library of Chemical Reactions
The Library of Chemical Reactions Currently, the reaction library contains 87 generic reactions – Reactions are defined in an ASCII formatted file and library can be easily augmented “A Collection of Robust Organic Synthesis Reactions for in silico Molecule Design” J. Chem. Inf. Model. 2011, 51, 309
Slide 17: The Library of Default Reactants
The Library of Default Reactants Our focus has been to assemble a “smallest, most diverse” library of reactants for lead hopping. Goal is to get the essence of an idea, not fully optimize it. But - Reactant library can be easily augmented. • The default reactant library consists of approximately 20,
Slide 18: Needle in the Haystack Validation
Needle in the Haystack Validation © Copyright 2014 Certara, L.P. All rights reserved.
Slide 19: Validation: Needle in the Haystack Experiment
Validation Needle in the Haystack Experiment Assess ability to find known drugs: – If we have the necessary reactions – And we have the necessary reagents – And if the search algorithm is sufficiently exploring chemistry space – Guided by a suitable scoring function – Given a known drug, do we: • Fi
Slide 20: How much of "Known" Drug Space do we cover?
How much of “Known” Drug Space do we cover? Examined Approved Drugs from Drug Bank – Contain only C, H, N, O, S, F, Cl, Br, I, or B – 30 < MW < 300 – Data Compiled data by therapeutic – No More than 3 aromatic rings – Mac Ring size less than 8 © Copyright 2014 Certara, L.P. All rights reserved. 20
Slide 21: Golden Needle Scoring Function
Golden Needle Scoring Function Score based on the distance of invented compounds to a reference (the known drug) using a vector of simple properties and 2D fingerprint similarities 2 𝑟𝑒𝑓𝑖 − 𝑚𝑜𝑙𝑖 2 max(𝑟𝑒𝑓𝑖 , 𝑚𝑜𝑙𝑖 )2 𝑖 ref – Reference Molecule mol – invented molecule to be scored…
Slide 22: Coverage of Small Molecule Drugs from DrugBank
Coverage of Small Molecule Drugs from DrugBank Getting Close Enough to get the Idea? © Copyright 2014 Certara, L.P. All rights reserved. 22
Slide 23: Coverage by Class
Coverage by Class Good coverage across most therapeutic classes Nuclear Hormone Receptors Ligands excluded by Filters Want to compare virtual syntheses with actual syntheses © Copyright 2014 Certara, L.P. All rights reserved. 23
Slide 24: Virtual Synthetic Pathway of Darunavir
Virtual Synthetic Pathway of Darunavir © Copyright 2014 Certara, L.P. All rights reserved. 24
Slide 25: Comments about Virtual Synthesis Schemes
Comments about Virtual Synthesis Schemes © Copyright 2014 Certara, L.P. All rights reserved.
Slide 26: Man vs. Machine - How well can you do?
Man vs. Machine - How well can you do? © Copyright 2014 Certara, L.P. All rights reserved.
Slide 27: Question 1: Muse Invent or Literature?
Question 1: Muse Invent or Literature? © Copyright 2014 Certara, L.P. All rights reserved. 27
Slide 28: Question 2: Muse Invent or Literature?
Question 2: Muse Invent or Literature? © Copyright 2014 Certara, L.P. All rights reserved. 28
Slide 29: Question 3: Muse Invent or Literature?
Question 3: Muse Invent or Literature? © Copyright 2014 Certara, L.P. All rights reserved. 29
Slide 30: Scheme 1 – Muse Invent
Scheme 1 – Muse Invent © Copyright 2014 Certara, L.P. All rights reserved. 30
Slide 31: Scheme 2 – Muse Invent
Scheme 2 – Muse Invent © Copyright 2014 Certara, L.P. All rights reserved. 31
Slide 32: Scheme 3 – Literature
Scheme 3 – Literature © Copyright 2014 Certara, L.P. All rights reserved. 32
Slide 33: Man vs Machine Results from ACS
Man vs Machine Results from ACS Chemists were given 4 reaction schemes – Each reaction scheme showed the synthesis path to a potential drug candidate. – For each reaction scheme, the chemists were asked whether the scheme was from the literature or from Muse Invent Conclusion: – The chemists surveye
Slide 34: Virtual Synthesis Pathways Comparable to "Real Chemistry"
Virtual Synthesis Pathways Comparable to “Real Chemistry” Chemists were asked to rate a set of 18 Synthesis Pathways – 10 of the Synthesis Pathways were selected from a set generated by Muse Invent – 8 were taken from literature synthesis of known drugs – Scale of 1 to 10 • 1= don’t see any way to m
Slide 35: What chemists have to say about Virtual Synthetic Pathways
What chemists have to say about Virtual Synthetic Pathways • I have a high degree of confidence the reaction schemes would work • I like that Muse performed reactions that joined two reactants and formed a ring in one operation. That builds a lot of complexity quickly. • Again, you’re making a ring
Slide 36: Summary
Summary © Copyright 2014 Certara, L.P. All rights reserved.
Slide 37: Summary
Summary Reaction Driven Molecular Evolution – Is a useful tool for multi-criteria drug design and generating ideas that meet multiple criteria – Using a library of 87 generic reactions and ~20,000 small diverse reactants, we are able to “rediscover” ideas similar to about a half of known small molec
Slide 38: Muse Invent Science
Muse Invent Science Publications: Automated generation of turn mimetics: Proof of concept study for the MC4 receptor. J.C. Baber, R. Lowe, J. Saunders, M. Feher, Bioorg. & Med. Chemistry, 2012, 20(11), 3565-74 Atom Coloring for Chemical Interpretation and De Novo Design for Molecular Design. K. Hase
Slide 39: Contacts & Follow-Up: Thank you!
Contacts & Follow-Up: Thank you! Americas Telephone (US): 888-708-7444 Telephone (Intl): +1-919-852-4685 Email: sales@certara.com Europe Telephone: +33 1 56 95 17 10 Email: sales@certara.com India Telephone: +91 984 500 6625 Email: sales@certara.com Japan Telephone: +81 (0)50-3786-9898 Email: sales@