Dispensing Processes Profoundly Impact Biological, Computational and Statistical Analyses Sean Ekins 1, Joe Olechno 2 Antony J. Williams 3 1 Collaborations in Chemistry, Fuquay Varina, NC. 2 Labcyte Inc, Sunnyvale, CA. 3 Royal Society of Chemistry, Wake Forest, NC. Disclaimer: SE and AJW have no affiliation with Labcyte and have not been engaged as consultants

“If I have seen further than others, it is by standing upon the shoulders of giants.” Isaac Newton Where do scientists get chemistry/ biology data?  Databases  Patents  Papers  Your own lab  Collaborators  Some or all of the above?  What is common to all? – quality issues

Data can be found – but …..drug structure quality is important  More groups doing in silico repositioning  Target-based or ligand-based  Network and systems biology  integrating or using sets of FDA drugs..if the structures are incorrect predictions will be too..  Need a definitive set of FDA approved drugs with correct structures  Also linkage between in vitro data & clinical data

Structure Quality Issues NPC Browser Database released and within days 100’s of errors found in structures DDT, 16: (2011) Science Translational Medicine 2011 DDT 17: (2012)

DDT editorial Dec This editorial led to the current work Its not just structure quality we need to worry about

Southan et al., DDT, 18: (2013) Finding structures of Pharma molecules is hard NCATS and MRC made molecule identifiers from pharmas available with no structures

How do you move a liquid? Images courtesy of Bing, Tecan McDonald et al., Science 2008, 322, 917. Belaiche et al., Clin Chem 2009, 55, Plastic leaching

 Extremely precise  Extremely accurate  Rapid  Auto-calibrating  Completely touchless  No cross- contamination  No leachates  No binding Moving Liquids with sound: Acoustic Droplet Ejection (ADE) Acoustic energy expels droplets without physical contact 8 Images courtesy of Labcyte Inc.

Using literature data from different dispensing methods to generate computational models Few molecule structures and corresponding datasets are public Using data from 2 AstraZeneca patents – Tyrosine kinase EphB4 pharmacophores (Accelrys Discovery Studio) were developed using data for 14 compounds IC 50 determined using different dispensing methods Analyzed correlation with simple descriptors (SAS JMP) Calculated LogP correlation with log IC 50 data for acoustic dispensing (r 2 = 0.34, p < 0.05, N = 14) Barlaam, B. C.; Ducray, R., WO 2009/ A1, 2009 Barlaam, B. C.; Ducray, R.; Kettle, J. G., US 7,718,653 B2, 2010

14 compounds with structures and IC 50 data. Barlaam, B. C.; Ducray, R., WO 2009/ A1, 2009 Barlaam, B. C.; Ducray, R.; Kettle, J. G., US 7,718,653 B2, 2010

A graph of the log IC 50 values for tip-based serial dilution and dispensing versus acoustic dispensing with direct dilution shows a poor correlation between techniques (R 2 = 0.246). acoustic technique always gave a more potent IC 50 value

12 14 Structures with Data Acoustic Model Tip-based Model Generate pharmacophore models for EphB4 receptor Acoustic Model Tip-based Model Test models against new data Acoustic Model Tip-based Model Test models against X-ray crystal structure pharmacophores Results Independent crystallography data Bioorg Med Chem Lett 18:2776; 18:5717; 20:6242; 21:2207 Independent data set of 12 WO2008/ Initial data set of 14 WO2009/010794, US 7,718,653 Experimental Process

Hydrophobic features (HPF) Hydrogen bond acceptor (HBA) Hydrogen bond donor (HBD) Observed vs. predicted IC 50 r Acoustic mediated process Tip-based process Ekins et al., PLOSONE, In press AcousticTip based Tyrosine kinase EphB4 Pharmacophores Generated with Discovery Studio (Accelrys) Cyan = hydrophobic Green = hydrogen bond acceptor Purple = hydrogen bond donor Each model shows most potent molecule mapping

An additional 12 compounds from AstraZeneca Barlaam, B. C.; Ducray, R., WO 2008/ A1, of these compounds had data for tip based dispensing and 2 for acoustic dispensing Calculated LogP and logD showed low but statistically significant correlations with tip based dispensing (r 2 = 0.39 p < 0.05 and 0.24 p < 0.05, N = 36) Used as a test set for pharmacophores The two compounds analyzed with acoustic liquid handling were predicted in the top 3 using the ‘acoustic’ pharmacophore The ‘Tip-based’ pharmacophore failed to rank the retrieved compounds correctly Test set evaluation of pharmacophores

Automated receptor-ligand pharmacophore generation method Pharmacophores for the tyrosine kinase EphB4 generated from crystal structures in the protein data bank PDB using Discovery Studio version Cyan = hydrophobic Green = hydrogen bond acceptor Purple = hydrogen bond donor Grey = excluded volumes Each model shows most potent molecule mapping Bioorg Med Chem Lett 2010, 20, Bioorg Med Chem Lett 2008, 18, Bioorg Med Chem Lett 2008, 18, Bioorg Med Chem Lett 2011, 21,

In the absence of structural data, pharmacophores and other computational and statistical models are used to guide medicinal chemistry in early drug discovery. Our findings suggest acoustic dispensing methods could improve HTS results and avoid the development of misleading computational models and statistical relationships. Automated pharmacophores are closer to pharmacophore generated with acoustic data – all have hydrophobic features – missing from Tip- based pharmacophore model Importance of hydrophobicity seen with logP correlation and crystal structure interactions Public databases should annotate this meta-data alongside biological data points, to create larger datasets for comparing different computational methods. Summary

Acoustic vs. Tip-based Transfers Adapted from Spicer et al., Presentation at Drug Discovery Technology, Boston, MA, August 2005 Adapted from Wingfield. Presentation at ELRIG2012, Manchester, UK NOTE DIFFERENT ORIENTATION Adapted from Wingfield et al., Amer. Drug Disco. 2007, 3(3):24 Aqueous % Inhibition Acoustic % Inhibition Serial dilution IC 50 μM Acoustic IC 50 μM Serial dilution IC 50 μM Acoustic IC 50 μM Log IC 50 acoustic Log IC 50 tips Data in this presentation No Previous Analysis of molecule properties

Strengths and Weaknesses Small dataset size – focused on one compound series No previous publication describing how data quality can be impacted by dispensing and how this in turn affects computational models and downstream decision making. No comparison of pharmacophores generated from acoustic dispensing and tip-based dispensing. No previous comparison of pharmacophores generated from in vitro data with pharmacophores automatically generated from X-ray crystal conformations of inhibitors. Severely limited by number of structures in public domain with data in both systems Reluctance of many to accept that this could be an issue Ekins et al., PLOSONE, In press

The stuff of nightmares?  How much of the data in databases is generated by tip based serial dilution methods  How much is erroneous  Do we have to start again?  How does it affect all subsequent science – data mining etc  Does it impact Pharmas productivity?

Simple Rules for licensing “open” data Williams, Wilbanks and Ekins. PLoS Comput Biol 8(9): e , : NIH and other international scientific funding bodies should mandate …open accessibility for all data generated by publicly funded research immediately Could data ‘open accessibility’ equal ‘Disruption’ Ekins, Waller, Bradley, Clark and Williams. DDT, 18:265-71, 2013 As we see a future of increased database integration the licensing of the data may be a hurdle that hampers progress and usability.

You can find CDD Booth 205 PAPER ID: PAPER TITLE: “Dispensing processes profoundly impact biological assays and computational and statistical analyses” April 8 th 8.35am Room 349 PAPER ID: PAPER TITLE: “Enhancing High Throughput Screening For Mycobacterium tuberculosis Drug Discovery Using Bayesian Models” April 9 th 1.30pm Room 353 PAPER ID: PAPER TITLE: “Navigating between patents, papers, abstracts and databases using public sources and tools” April 9 th 3.50pm Room 350 PAPER ID: PAPER TITLE: “TB Mobile: Appifying Data on Anti-tuberculosis Molecule Targets” April 10 th 8.30am Room 357 PAPER ID: PAPER TITLE: “Challenges and recommendations for obtaining chemical structures of industry-provided repurposing candidates” April 10 th 10.20am Room 350 PAPER ID: PAPER TITLE: “Dual-event machine learning models to accelerate drug discovery” April 10 th 3.05 pm Room 350