Transcript Technology Integration for Analysis of High Throughput Cellular

Technology Integration for Analysis of High
Content/Throughput Cellular Data: The
Cytomics Approach
J. Paul Robinson
Professor of Immunopharmacology &
Professor of Biomedical Engineering
Purdue University
This presentation will discuss current ideas for analysis of
live cell data incorporating multivariate approaches. It will
outline the major problems faced by present generation
technologies and provide insight into future advances. Key
to the success of future technologies will be an
understanding of informatics and high-speed data
processing including advanced image analysis.
www.cyto.purdue.edu
Goals of this Presentation



Introduce Cytomics
Identify current & forthcoming
issues & technologies
Call attention to issues that need to
be addressed
Note: Added for the web version of this presentation:
These slides were perfect!! In Powerpoint. However, as amazing
as it might seem, the Powerpoint web converter is pretty much
the usual Microsoft Disaster Product (MDP).
So, many animations and boxes with text look strange and the
text often fails to remain inside the boxes, lines are everwhere….
What is the Cytomics Approach?



Discovering the functional relationships between the
cell (Cytome) and the metabolic pathways
(Proteomics-proteome) resulting from genetic control
mechanisms (Genomics-genome) –
Some relate Cytomics to what is being termed
functional genomics.
By definition we are expanding the information being
collected in every system because we also want
functional data, not just morphological, phenotypic or
genotypic.
..the cell is the ultimate functional
endpoint…



Cytomics is going to be important because it is the
cell that is the ultimate functional endpoint. The cell is
the minimal functional unit within our physiology
and thus the functional unit that can be manipulated.
Complexity of cell function is only part of why
Cytomics will become a major field of study. Every cell
is different. By studying each cell's unique function,
that cell type can be further modeled for subsequent
analysis using statistical techniques.
As the field of tissue engineering explodes, it will not
be long before cellular engineering will be a most
important component of which an essential element
will be a full understanding of Cytomics.
Cytomic-realignment….


Within a short time, no pharmaceutical
company will operate without encompassing
the essential features of Cytomics
Drugs design will operate at the level of
modified cellular functions, cytome-alignment
or cytomic-realignment will become the
"cellular form" of tissue engineering.
..how does the cell operate…




This knowledge will require a better-thanever understanding of how the cell operates,
how to measure cell function, and how to
characterize the live cell in minute detail.
Single-cell analysis techniques will become
enhanced and exquisitely sensitive.
New technologies must be developed and
new analytical tools will be required to
extract these new data.
Of these analytical tools, informatics will
continue to play a crucial role in cell biology.
The big link…



Cytomics links technology to
functional biology
Cytomics relates measurement &
detection to structure & function
Cytomics integrates tools like flow
cytometry, image cytometry, etc.
with proteomics.
Cytomics….summary so far





Integration of technologies
Functional role of system components
Relates measurement & detection to
structure & function
Brings together traditional cytometry
and non traditional cytometry
Informatics now assumes a primary
rather than a secondary role in
cytomics
Current Emphasis
protein
gene
cell
Hey buddy…
Don’t you
know you
genes,
proteins and
organelles are
in my territory
now!!
protein
cell
Live
Cell
gene
cell
Please wait till slide animation is complete
Slide animation
is complete
Systems Integration

Analytical Cytology




Image Cytometry & Analysis





Flow cytometry
Single cell analysis systems
Tissue analysis (after cell separation)
Single cells
Tissues and sections
Cell culture systems
3D and 4D cell culture environments
Proteomics


Proteins from specific cell populations
Rapid identification
Imaging Technologies?
• DNA arrays
• “Quantitative” fluorescence assays
• High Throughput assays (96-384-1536 well plates)
• Elispot
• Drug effect assays, Cyto-toxicity
• Toxicology assays
• Apoptosis
• Cell proliferation, Cell ploidy
• Cytoplasm-to-nucleus transportation
• Hormone receptors, Growth factors
• Gene amplification or deletion, Gene fusion
• Chromosome imbalances
• And the list goes on……..
Next Generation Instruments…











40 fluorescent colors (40-50 variables & 100-200
parameters)
Lots of other spatial measurements
Lifetime
Hyperspectral Imaging - Spectral
unmixing/deconvolution technologies
Multiple probe systems
Complex analytical tools – informatics approaches
120,000 events/s for flow systems (4 x 108/hour)
Very high speed for imaging systems
Permanent and accurate alignment
Intelligent interfaces and operating systems
Direct links with diagnostic expert systems
This is HIGH Content
• Huge number of
Variable &
Parameters
• Very High Speed
Rapid
Identification
or
Diagnostics
• Huge data sets
• Opportunity for
Rapid classification
Much of this can become
Real-Time decision making
Laser Scanning Cytometer





First “modern”
high content static
cell analysis
system
Very high content
Moderate speed
Very high data
storage required
Data-base friendly
Concept first published: Kamentsky & Kamentsky, Cytometry 12:381-7, 1991
Many Spectra in Flow Cytometry
Roederer, et al
Multi-Component Systems
http://www.amnis.com
Amnis Corp
Slide animation
is complete
Future integration
Cell Sorter
Eprogen-Beckman-Coulter automated
protein separation system
ProteomeLab™ PF 2D Protein
Fractionation System from
Beckman-Coulter
http://www.beckman.com/products/instrument/protein/proteomelab_pf2d_dcr.asp
High through-put flow cytometry - Multiplexing
High Through-Put Flow Cytometry
Dr. Larry Sklar, Cytometry 44:83-90 (2001)
Multispectral microscopy – Not more colors!!!
Greyscale
image
Color image
Expansion/rebirth of the Landsat Concept from the 1970s
Multispectral
image
Multispectral microscopy
Intensified
camera
Intensified camera
CCD camera
AOTF
Camera controller
AOTF
AOTF controller
Monitor
Microscope controller
Microscope
PC computer
Purdue Spectral
Imaging Project
Enabling Technology: Liquid tunable filters
Single
bandpass
Measured center wavelength (nm)
Lyot filter
(static)
750
700
650
600
550
500
450
400
400
450
500
550
600
650
700
750
Wavelength “dialed-in”
LCTF (randomly tunable)
High precision and accuracy
Slide from Dr. Richard Levenson, CRi, Inc.,35B Cabot Rd.,Woburn, MA 01801, www.cri-inc.com
High-resolution cytology segmentation
NOTE: this slide has animation – you should wait till it finishes
Characteristic
Spectra
Conventional
RGB Image
Spectrally
segmented Image
Slide animation
is complete
Wavelength (nm)
High spectral resolution increases utility of spectrally responsive indicator dyes
Slide from Dr. Richard Levenson, CRi, Inc.,35B Cabot Rd.,Woburn, MA 01801, www.cri-inc.com
Nuance-Micro
Slide from Dr. Richard Levenson, CRi, Inc.,35B Cabot Rd.,Woburn, MA 01801, www.cri-inc.com
Multispectral Imaging – Zeiss Meta
Ability to select a range of wavelengths
As desired by the user
Visualization of morphology of cells embedded within a
collagen matrix
NOTE: this slide has animation – you should wait till it finishes
Slide animation
is complete
Publications: http://www.cyto.purdue.edu/flowcyt/research/pub1.htm
Small Intestinal submucosa – BSL-based
visualization
Publications: http://www.cyto.purdue.edu/flowcyt/research/pub1.htm
Visualization of collagen matrices — laser scanning confocal
microscopy using backscattered light
NOTE: this slide has animation – you should wait till it finishes
Slide animation
is complete
Combinatorial based classification using
multivariate analysis
Robinson et al - Cytometry 12:82-90, 1991
Modern optical microscopy
NOTE: this slide has animation – you should wait till it finishes

Confocal microscopy




Multiphoton microscopy




UV+VIS Fluorescence
Backscattered laser light
Environmental Control
2-p and 3-p fluorescence
SHG
Lifetime (B&H)
Multispectral microscopy

Wide-field
Confocal (Bio-Rad
Rainbow, Zeiss Meta)
Purdue “Spectralfluor”

CRI’s Nuance-Micro


Core technologies
Slide animation
is complete
So what does the future look like
for data processing?





It’s moving fast!
We are pushing multiple
technologies simultaneously
Data processing is well beyond
human capacity – Informatics
Functional studies bring
exponential complexity
Real-Time decision making will
be the next requirement
Is there life after HCS





Fortunately HCS is not eternal
We are going through an evolution of
rapid technology change
We are trying to use current HCS
technologies to do everything – that will
change
We all do need to be cautious – some
companies will develop great technologies
and then go broke! What will you do
then?
The only “dream machine” is in a dream
Acknowledgements

Bio-Engineering








Microbiology/Biofilms



Bartek Rajwa
Jennie Sturgis
Wamiq Ahmed
Muru Venkatapathi
Silas Leavesley
Jim Jones
Padma Varadharaajan
Stephanie Sincock
Gerald Gregori
Cytomics




Jia Lu
Kathy Ragheb
Cheryl Holdman
Gretchen Lawler
Additional Material for Discussion

The following materials were
incorporated to highlight a number
of educational facilities.
Appendix
Some Key Web Links
www.cyto.purdue.edu
http://www.cyto.purdue.edu/HCS
Issues that can be addressed:
1.
2.
3.
4.
5.
Discussion Group - Communication
Educational – knowledge development and training
Issues of data file standards
Issues of calibration standards
Image processing issues – algorithms and processes
Opportunity?
Discovery Park, Purdue University
Center for Applied Cytomics
Expanded educational role
Can we partner with HCS users and developers
Can we train in the basic issues
ISAC






International Society for Analytical
Cytology
Forum to address issues
Cell-focused – understands “high
content”
Has an existing structure
International meeting May 23-28, 2004
in Montpellier, France
www.isac-net.org
www.cyto.purdue.edu/HCS
EMAIL me suggested questions
[email protected]
Survey
of
HCS Users
Site for Lectures and Presentations
Just under a ton of educational materials!
http://www.cyto.purdue.edu/flowcyt/educate/pptslide.htm
Data standards – Conversion to standard



LData
Converts any
instruments
specific file to
FCS for
software
analysis
Can be used
as
independent
utility or
included in
code
Image Software reviews

Image
Analysis
Software Tutorials



Dr. Gerald
Gregori set of
tutorials
More software
reviews coming
Independent
evaluation