This presentation introduces drug discovery, robotics and laboratory automation at Scripps Florida. It also examines a case study of the invention of a novel machine vision instrument which has recently been approved as a U.S. Patent and licensed for commercialization. Lessons learned throughout the process of prototyping, development, and commercialization of this instrument will be presented at the conclusion of the talk.

1. Robotics and
Laboratory Automation
at Scripps Florida
Pierre Baillargeon

2. About Scripps Florida
I.
Introduction to drug discovery, robotics and laboratory
automation at Scripps Florida
II. Case study: the development of a novel machine
vision instrument
III. Lessons learned along the way
The Scripps Research Institute © 2014 – All rights reserved

3. The Lead Identification Laboratories at Scripps
Where robotics, chemistry and biology join forces to help discover new drugs
Integrated Assay Development, uHTS and Compound Management
• Located in Jupiter, FL
• >200 targets screened in over 235 robotic experiment
campaigns to generate more than 80 million data points for
academic & industrial collaborators
• Multiple Sclerosis (S1P1) drug candidate entered phase
2/3 clinical trials in October 2012
• >982K Compound Screening Collection
• >622K Proprietary (largest in
academia, ~30k unique compounds, focused
sub-libraries, professionally curated)
• >360K Public Domain (NIH)
The Scripps Research Institute © 2014 – All rights reserved

4. Introduction to Drug Discovery
Average time to develop a new medicine = 10-15 years
The Scripps Research Institute © 2014 – All rights reserved

5. Introduction to Drug Discovery
How can we shrink the timeline and reduce the cost of drug
discovery?
The Scripps Research Institute © 2014 – All rights reserved

6. Introduction to Drug Discovery
1 experiment per test tube
Up to 1,536 experiments per microtiter plate
Using laboratory automation reduces unit cost & increases
throughput
The Scripps Research Institute © 2014 – All rights reserved

7. Introduction to Drug Discovery
The Scripps Research Institute © 2014 – All rights reserved

8. Introduction to Drug Discovery
Bulk reagent dispensing
Multichannel pipetting
Result Detection & Reporting
The Scripps Research Institute © 2014 – All rights reserved

9. Introduction to Drug Discovery
Software automation plays a large role in increased efficiency and quality control
1st generation
2nd generation
3rd generation
No automation
Software automation
Software automation + hardware automation
The Scripps Research Institute © 2014 – All rights reserved

10. Example of a typical process at Scripps
COMPOUND SUBMISSION
• Has the correct sample been put into the labware?
• Is the volume and concentration of the sample accurately recorded?
COMPOUND PROCESSING
• Has the sample store delivered the correct samples?
• Has the liquid handling automation pipetted the samples properly?
• Has the plate been registered correctly?
• Is the sample itself of high quality?
COMPOUND DELIVERY
• Do we have a record of what we are delivering?
Room for improvement! How can we QC samples more efficiently?
The Scripps Research Institute © 2014 – All rights reserved

11. Issues With Compound Libraries
I.
Introduction to drug discovery, robotics and laboratory
automation at Scripps Florida
II. Case study: the development of a novel machine
vision instrument
III. Lessons learned along the way
The Scripps Research Institute © 2014 – All rights reserved

12. Why Do We Need Plate Auditor?
Are empty/partially filled
wells in Quadrant 1?
My db says
this well is
empty. Does
it actually
have sample
in it?
The Scripps Research Institute © 2014 – All rights reserved
Do these wells
contain precipitate?

13. Issues With Compound Libraries
Should these wells really be empty? What
does my database say?
Do these wells
contain precipitate?
My experiment is sensitive to purple
compounds-where are they?
The Scripps Research Institute © 2014 – All rights reserved
Are these wells
partially filled?

14. What is Plate Auditor?
• A custom HTS plate reader, incorporating
recent advances in machine vision, image
analysis, & the spectroscopy sciences
• It saves labor by automatically identifying
& annotating issues specific to compound
libraries:
• Colored compounds
• Precipitate / Crystallization
• Low volume / Full wells
• It can be used in “stand-alone” mode, or
integrated with automation
• Works with microtiter plates
• Fast: reads a 384 well plate in <1 minute
• All measurements are non-contact
• User-friendly, intuitive analysis software
The Scripps Research Institute © 2014 – All rights reserved

15. Features: Instrument Functionality
The Scripps Research Institute © 2014 – All rights reserved

16. Taking an idea to market
The Scripps Research Institute © 2014 – All rights reserved

17. Evolution of hardware
Proof of concept
1st generation
2nd generation
3rd generation
The Scripps Research Institute © 2014 – All rights reserved
4th generation

18. Evolution of software
1st generation
2nd generation
The Scripps Research Institute © 2014 – All rights reserved
3rd generation

19. Implementation
How does it work?
The Scripps Research Institute © 2014 – All rights reserved

20. Before Processing
Features: Color Recognition & Classification
• Plate Auditor detects and classifies colored
compounds
• Plate Auditor identifies the “standard” colors
listed below:
Color
Detected
Dye Used
Limit of
Detection
(uM)
Allura Red
35.1
Brilliant Green
9.0
Blue
Euroglaucine
0.6
Yellow
After Processing
Red
Green
Tartrazine
12.7
Violet
Gentian violet
1.3
The Scripps Research Institute © 2014 – All rights reserved

21. Features: Precipitate Detection
Compound plates contain precipitates which are masked by colors in the
wells:
Plate Auditor Precipitate analysis is:
• Insensitive to color
• Able to detect ppt. beyond “naked eye” inspection
The Scripps Research Institute © 2014 – All rights reserved

22. Features: LIMS/Corporate Database Auditing
RESULTS OF COMPOUND PLATE QUERY
IN CORPORATE DATABASE:
• Scripps database (MDL Symyx/Plate Manager)
displays several wells as empty (white rectangles
in figure on left)
RESULTS OF THE SAME COMPOUND
PLATE QUERY IN Plate Auditor
DATABASE:
• Plate Auditor compares instrument results to
records archived in corporate db
• Plate Auditor identifies discrepancies with
corporate db records(red “x’s” in figure on left):
• Colored compounds (upper left)
• Full wells that corporate db assigns as “empty”
Plate Auditor CAN BE USED TO AUDIT/UPDATE CORPORATE DB RECORDS
The Scripps Research Institute © 2014 – All rights reserved

23. Opportunities For Error
What new information do we obtain with Plate Auditor in the process?
• We know which wells have been filled and which have low volume
• We know which wells contain colored samples
• We know which wells contain compounds that have precipitated
What can we do with this information?
• Compare filled & low volume wells against LIMS to verify proper liquid handling & plate
registration
• Compare „color fingerprint‟ of samples against database of known samples to identify issues with
incorrect concentration & samples
• Take measures to put precipitated samples back into solution (sonication, heating, etc.)
• Keep a historical record of the health of the plate at the point when it was delivered (a receipt of
transaction)
The Scripps Research Institute © 2014 – All rights reserved

24. Practical Examples: 384 Well Plate Timing
To visually inspect three 384 well
plates:
Plate #1 – 3 minutes 12 seconds
Plate #2 – 1 minute 46 seconds
Plate #3 – 1 minute 40 seconds
In the same amount of time, Plate
Auditor can automatically inspect
fifteen 384 well plates!
The Scripps Research Institute © 2014 – All rights reserved

25. Practical Examples: HTS Library Analysis
To date, Scripps has run over 1,700 plates through the Plate
Auditor, resulting in over 2,600,000 wells which have been
analyzed for color, precipitate and low volume artifacts.
How can we leverage all of this new data to improve our workflow?
The Scripps Research Institute © 2014 – All rights reserved

26. Practical Examples: Low volume detection
Did my pipettor transfer all compounds properly?
Plate Auditor enables just-in-time QA of compound management liquid handling
The Scripps Research Institute © 2014 – All rights reserved

27. Practical Examples: Color Analysis
Plate Auditor data can be used to monitor consistency of sample color over time…
A sample provided by an external source in 2009 vs. the
same sample provided by the same source in 2011.
With Plate Auditor, we can
automatically query and
identify inconsistencies
across entire compound
collections.
Is this the same sample? Is the concentration recorded
correctly?
The Scripps Research Institute © 2014 – All rights reserved

28. Practical Examples: Color Analysis
Why should we care about sample color?
The Scripps Research Institute © 2014 – All rights reserved

29. Practical Examples: Color Analysis
Running color comparison reveals
color inconsistencies…
Sample 1
Sample 2
Two copies of plate prepared by external vendor reveal
inconsistencies
Sample 3
Sample 4
Sample 5
The Scripps Research Institute © 2014 – All rights reserved

30. Practical Examples: LIMS Comparison
When checking reformatted plates, CM team noticed high # of LIMS disagreement wells…
… examining the plate detail view reveals a pattern in quadrant 1…
… checking the LIMS record reveals samples in quadrant 1 are not registered …
The Scripps Research Institute © 2014 – All rights reserved

31. Practical Examples: LIMS Comparison
… CM team updates LIMS record and runs plate through Plate
Auditor to confirm the corrected LIMS record matches the
physical locations of samples within the plate.
The Scripps Research Institute © 2014 – All rights reserved

32. Introduction to Plate Auditor
I.
Introduction to drug discovery, robotics and laboratory
automation at Scripps Florida
II. Case study: the development of a novel machine
vision instrument
III. Lessons learned along the way
The Scripps Research Institute © 2014 – All rights reserved

33. Don’t take no for an answer
•
•
Two grant applications were denied during the period of instrument
development
One year was spent unsuccessfully demonstrating the technology to various
companies looking for a commercial partner
If at first you don‟t succeed… put it in the back of your truck and
drive it to a trade show (but be ready to hustle)
The Scripps Research Institute © 2014 – All rights reserved

34. Keep an eye open for disruptive technologies
3d printing example:
Drosophila flies used in Neuroscience research at
Scripps & FAU
•
•
Fixed design, if it doesn’t
work as desired there is
little room for improvement
~$600 for device
•
•
Custom design, enables fast
iteration, new prototypes
fabricated in hours
~ $1 for plastic, plexiglass
cover, ~$50 for USB
webcam
The Scripps Research Institute © 2014 – All rights reserved

35. Keep your eyes open for new applications
The Plate Auditor was also helpful in
documenting QC failures at the factory
where our injected molded labware is
manufactured
The Scripps Research Institute © 2014 – All rights reserved

36. Get ready to be a technology evangelist
If you invent or make something, be prepared to talk about it
The Scripps Research Institute © 2014 – All rights reserved

37. Documenting your work
“If you don’t document it, it doesn’t exist”
http://makezine.com/2013/09/12/if-you-dont-document-it-it-doesnt-exist/
Show your projects to the world! Where to show
your projects IRL:
• Local Meetup groups (meetup.com)
• Maker Faires
• Your local makerspace
Reasons to document your projects and work:
• Immensely useful for job interviews
• Interacting with communities is rewarding (and
often fun)
• To have something to talk about when you get
invited to speak at a FAU Engineering seminar
Where to document your projects and work online:
• LinkedIn, Reddit, Facebook, Twitter, Instagram, et
c, personal
blog, Instructables, Hackaday, Makezine, etc.
The Scripps Research Institute © 2014 – All rights reserved

38. Growth of robotics
According to a February 2014 report from Littler Mendelson (largest
U.S. based labor & employment law firm devoted to representing
management in employment): “Robotics is the fastest growing industry
in the world, poised to become the largest in the next decade. The
increasing use of artificial intelligence (AI) and related technologies in
the workplace are dramatically changing the employment landscape.”
http://www.jdsupra.com/legalnews/the-transformation-of-theworkplace-thro-67856/
The Scripps Research Institute © 2014 – All rights reserved

39. P.S. Don’t forget to give back to the community
2013 summer Maker Camp
Scripps ongoing education outreach
efforts
The Scripps Research Institute © 2014 – All rights reserved

40. Acknowledgements
Scripps Florida
Funding Corporation
http://hts.florida.scripps.edu
Lead Identification:
Pierre Baillargeon
Peter Chase
Ross Einsteder
Anita Feng
Virneliz Fernandez-Vega
Lina DeLuca
Franck Madoux
Laura Pedro-Rosa
Louis Scampavia
Timothy Spicer
The Scripps Research Institute © 2014 – All rights reserved
bpierre@scripps.edu