Overview of the Biotech Industry

1
Overview of the Biotech Industry

• Srinivasan Seshadri, CEO, Strand Genomics

2
EMERGING DISCOVERY PROCESS
The drug discovery process is currently being
transformed by emerging technologies
Basic process
Current impact of new technologies
Fundamentally new approach
Batter/faster
Better/faster
No major breakthrough

• Old world
• Molecular biology
• Physiology
• Biochemistry
• New world
• Genomics
• Combinatorial chemistry
• High throughput screening

• Still reliant on in vitro/in vivo models of
  disease
• Greater use of more sophisticated genetic
  models, but currently complex and slow

• Old world
• Slow, largely manual chemical synthesis of leads
• Slow, manual screening on limited range of assays
• New world
• More sophisticated/automated (high throughput)
  screening has increased lead identification
  productivity by 30 times
• Rate of compound generation increased by factor
  gt1,000 through combinatorial chemistry techniques
  (although not clear what are useful)

3
GENOMICS A BIOLOGICAL DEFINITION
Genomics is central to this evolving landscape.
The goal of Genomics is to unravel the genetic
basis of health and disease providing a huge
array of potential drug targets. Given the
complexity of the genome and the volumes of data
being generated, significant challenges exist in
accessing and leveraging this data effectively.
New IT based arenas are emerging to do this

• Genomics is the study of the genetic composition
  of an organism and provides information on the
  structure, role and genetic linkage of genes.
  Some gene function is implicated in disease and
  it is therefore believed that better, more
  specific information about the origins of a
  disease will lead to more effective treatments.

AT
TA
GC
CG
CG
GC
AT
TA
The characteristics (phenotype) of each
individual. . .
. . . and their organs and tissues. . .
. . . are determined by their genetic makeup
(genotype). Every cell contains the full
complement of individuals genetic material the
genome
The genome consists of a length of double
standard DNA to which are attached 4 types of
molecule of bases. There are 3 billion of these
in total
Some of these bases code for proteins the cell
manufactures protein using the DNA template).
Others fill in the gaps and have no other
function. A gene represents a section of DNA
which codes for a protein or other functional
piece of cellular machinery (e.g., tRNA, rRNA)
Genomics represents a paradigm shift in disease
treatment from underlying mechanism to root
cause CSO,
Genomic-co
4
BIOINFORMATICS A DEFINITION
Bioinformatics describes one of two information
driven arenas within pharmaceutical drug
discovery. . .
Focus on this document
Description
Key applications
Relevance for pharmacos

• Information technology designed/used to generate
  and access genetic data and derive information
  from it.

• Searching external genomic databases
• Constructing and managing proprietary databases
• Extracting information from data
• Gene expression in health and disease
• Gene function in health and disease

• Pharmacos need to be able to effectively access
  external sources
• Pharmacos need to create proprietary databases
  (derived from data from multiple sources) so that
  they can be tailored to the needs of internal
  discovery function
• Targets need to be identified and their role
  defined leveraging genetic data

Bioinformatics
Informatics

• Information technology used to design molecular
  libraries to interact with identified targets

• Molecular structure design
• Structure-activity relationships
• Molecular library management/manipulation

• IT solutions required to manage the increasing
  scale of molecule generation with discovery
  process
• Predominantly addressed within pharmacos although
  may require leveraging links with partners and
  across multiple geographies

Cheminformatics
5
INFORMATICS USED IN DRUG DISCOVERY
. . . and currently assists in leveraging genetic
data. As the arena develops, the current
boundaries with cheminformatics are likely to blur

• Analysing sequence data is just the starting
  point for bioinformatics the key step will be
  relating that data back to protein structure
• J. Craig Venter, TIGR

Identify lead compounds which interact with
targets
Define target
Target role
Optimise leads
Bioinformatics
Cheminformatics
Activity

• Trawl genomic databases for genes of interest

• Define amino acid sequence derived from gene

• Determine structure of protein and how it folds
  into active molecule

• Define protein activity
• Define likely molecule structures to interact
  with target

• Trawl molecular databases for likely activity
  against target

• Refine search/ development of lead compound

• Links with combinational chemistry and high
  throughput screening

6
KEY TECHNOLOGICAL AND SCIENTIFIC
HURDLES/CHALLENGES
Many significant hurdles remain before the
value of bioinformatics can be fully exploited
High
Medium
Low
Key challenges
Why important
Size of challenge

• Developing tools which improve human-computer
  interface
• Allowing disparate systems to interface with
  genomic databases
• Developing industry standard low cost
  infrastructure to access databases (internal and
  external)
• Increasing the efficiency and effectiveness of
  database mining

• Lower the barriers for effective use of computers
  by multiple disciplines to access databases and
  translate data into user-friendly format
• IT architectural differences constrain access to
  databases
• Developing proprietary infrastructure is often
  too costly/time consuming
• Current database data mining generates vast
  quantities of irrelevant to search criteria

Technology/ IT-based
Key challenges for bioinformatics

• Predicting tertiary protein structure currently
  carried out using laborious methodologies of
  moderate efficacy e.g., X-ray crystallography
• Predicting protein function currently expensive
  and time consuming
• Understanding how genes and proteins are
  expressed/modified in vivo currently unknown

• Most drug targets are proteins
• Need to have clear understanding of role of
  protein to drug design
• Experimental methodologies being developed e.g.
  gene knockouts, although time consuming and
  ill-developed
• Gene structure predicted from genetic sequences
  may not reflect the gene expressed in vivo, and
  proteins can be modified into alternative
  structures with differing function to those
  predicted

Science-based
Source interviews, press search
7
BIOINFORMATICS HIERARCHY OF POSSIBILITIES
The current activity is only a small part of what
bioinformatics (integrating with the other
emerging technologies) could contribute to the
way we understand and treat disease
Examples
Status

• Replacing animal-based wet biology with
  computer-based predictive models

• Very preliminary

• Insilica research

• Replacing crystallography to determine protein
  structure with predictive models
• Generating better information about disease and
  patient populations allowing better targeting of
  clinical trials

• In early development
• In medium stage development

Increasing the productivity of discovery

• Mining genetic databases from normal and diseased
  populations to elucidate gene function
• Increasing the effectiveness/efficiency of
  generating information from genomic data

• Ongoing
• Ongoing

Increasing the effectiveness and efficiency of
genomic database mining
Source interviews, articles
8
BIOINFORMATICS INDUSTRY EVOLUTION A DESCRIPTION
To-date, bioinformatics has developed
symbiotically with Genomics. It is now emerging
as a field in its own right
Gene function-driven
Academia-driven
Genomics-driven

• Multiple academic groups leveraging existing IT
  competencies to develop insights into
  identification and role of genes

As genomic data becomes easier to generate,
genomic companies (positional cloners and
sequencers) develop IT systems to facilitate
access to genome sequences Key differentiating
factor is scope and scale of gene databases to
which clients have access
Focus of effort becomes role and function of
genes, and in particular, gene products.
Organisations develop IT skills to push the
boundaries of knowledge (e.g., predicting protein
structure-activity relationships). Key
differentiating factor is becoming a companys
ability to provide bioinformatic solutions to
extract information from genes
Source Team interviews
9
BIOINFORMATICS INDUSTRY EVOLUTION KEY MILESTONES
1981 First 579 human genes mapped
1983 Method for automated DNA sequencing
(Carruthers Hood)
Key Scientific Milestones
1992 First genetic linkage map of entire human
genome published, and first whole human
chromosome physical maps (Y and 21)
1983 Huntingdons disease gene demonstrated to be
on chromosome 4 (Gusella)
1991 Expressed sequenced tags (ESTs) created
(Venter)
1977 Chemical method for sequencing DNA devised
(Gilbert Maxam)
1972 first DNA cloning (Boyer Cohen
GENOMICS-DRIVEN
ACADEMIA-DRIVEN
GENE FUNCTION-DRIVEN
1996/7 Genomic industry broadening value
proposition
1990 Human Genome Project launched
Genomics/ Bioinformatics Industry Activity
1977 First genetic engineering company,
Genentech, founded
1982 Genbank established
1997 Emergence of bioinformatic players with no
genomic heritage e.g., - Pangea - MAG -
NetGenics
1993 Incyte goes public, the first of many U.S.
genomic companies to do so
1988 Human Genome Organisation (HUGO) founded
10
SUMMARY

• TECHNOLOGIES
• Three broad enabling technologies are driving
  progress in drug RD
• Genomics leads to better disease understanding
  and target identification
• Combinatorial chemistry generates more lead
  compounds
• High throughput screening tests more leads on a
  greater number of targets

• BIOINFORMATICS
• The explosion of data and the increasing demand
  for sophisticated analytical tools has given rise
  to a rapidly growing bioinformatics market with
  three major service areas
• Database providers who generate and organize
  genome and discovery data
• Discovery software providers who provide
  cutting-edge IT solutions to elements of the
  discovery process
• Research enterprise ASPs who integrate multiple
  databases and analysis tools into a single
  platform

11
THREE BROAD TECHNOLOGIES ARE DRIVING DRUG
DISCOVERY

• Study of both structural and functional aspects
  of the genome, including both genes and proteins,
  leading to a greater understanding of cellular
  processes and disease

GENOMICS
Supported by BIOINFORMATICS
HIGH THROUGHPUT SCREENING
CATALYTIC/ COMBINATORIAL CHEMISTRY

• Rapid and systematic generation of a variety of
  molecular entities, or building blocks, in many
  different or unique combinations

• Use of robotic automation to allow for massive
  parallel experimentation and testing of many
  compounds or targets

12
MANY SPECIFIC EMERGING TECHNOLOGIES HAVE LED TO
THE ADVANCES IN GENOMICS, COMBINATORIAL CHEMISTRY
AND HIGH THROUGHPUT SCREENING

• Antisense
• Transgenics
• Gene therapy
• Pathway mapping
• Surrogate markers
• Animal-free disease models
• Genetic networks

• HT DNA sequencing
• HT proteomics
• Biochip microarrays
• Pharmacogenomics
• Biosensors

• Synthetic biopolymers
• Biochemical drug delivery and encapsulation
  systems

• Lab Automation
• Micromachines/miniaturization

• Intelligent chemical systems

• CC library arrays
• Chemical chips
• Advanced biophysical assays

Note HT High Throughput CC Combinatorial
Chemistry
13

• TECHNOLOGIES
• Three broad enabling technologies are driving
  progress in drug RD
• Genomics leads to better disease understanding
  and target identification
• Combinatorial chemistry generates more lead
  compounds
• High throughput screening tests more leads on a
  greater number of targets

• BIOINFORMATICS
• The explosion of data and the increasing demand
  for sophisticated analytical tools has given rise
  to a rapidly growing bioinformatics market with
  three major service areas
• Database providers who generate and organize
  genome and discovery data
• Discovery software providers who provide
  cutting-edge IT solutions to elements of the
  discovery process
• Research enterprise ASPs who integrate multiple
  databases and analysis tools into a single
  platform

14
BIOINFORMATICS IS THE BRAINS OF BIOTECHNOLOGY
In order for Genomics, HTS, and combinatorial
chemistry to have impact, they must increasingly
rely on bioinformatic capabilities.
1 Science, Bioinformatics in the Information
Age April 2000 287 1221 Source Brains of
Biotechnology is from Karl Thiel, Biospace.com
15
NEW TECHNOLOGIES ARE DRIVING THE NEED FOR
BIOINFORMATICS DATA AND ANALYSIS CAPABILITIES
ANALYSIS TOOLS
EXPLOSION OF DATA

• Gene (DNA) sequences
• Protein sequences
• SNP mapping and disease mapping
• Gene expression profiles by tissue, species, and
  drug influence
• Protein expression profiles
• Proteinprotein interaction profiles
• Protein structure information
• CC libraries
• Screening activity data (SAR)
• Toxicology databases

• Sequence alignment searches (BLAST)
• Relational alignment programs (phylogeny)
• Virtual lab processes software (PCR, elongation)
• Protein folding algorithms
• Structure-based target design using virtual SAR
  modeling
• Virtual CC generation and screening
• ADME and toxicology profiling software

NEW TECHNOLOGIES
Demand for discovery software
Demand for different types of databases
16
Growth in Global Bioinformatics
The global market for bioinformatics is expected
to show significant growth over the next five
years. However the state of infancy of the
industry poses credibility issues on the
estimates from research houses
10-20Bln
5Bln

• Numbers likely to include
• Software solutions
• Automations tools
• Hardware such as microarrays

300m
1998
2003
17
Growth in Indian Biotechnology
The current biotechnology market in India is
focused on the AgBio, Industrial and Vaccine
sectors, but will see emerging opportunities in
Bioinformatics and vaccines
100??
Bioinformatics
Genome Technologies
Vaccines
100500m
The future will witness additional opportunities
in Informatics and related genome based
technologies
22
Industrial
25
Ag Products
Health Products
47
1998
2003
Source Biosupportinida
18
Pharmaceutical RD Budgets
Projected growth in Pharma and Biotech RD
spending will enable the industry to attain its
projected targets
100100Bln
46B
20B
13B
13B
Typical IT budgets will be 10-20 of total RD
7B
Discovery
Production/ manufacturing
PreClinical
Clinical Trials
Clinical CMC
Source PhRMa
19
Summary of Findings
The Lehman Report consisted of interviews with
decision makers in Pharma and Biotech and
highlighted some interesting findings

• New Biology will significantly increase RD
  costs- a large chunk of which are technology
  driven
• Companies will see substantial pressure on
  earnings
• Attempts to use today's relatively immature
  technology will result in higher failure rates
  amongst
• novel targets. These failures will likely also
  stretch out the time period for the arrival of
  new drugs that
• Genomics promises

• High risk of novel target failure
• Less understood (only 8 publications per novel
  target vs. 100 for those generated by
• conventional methods
• Companies pushing these less understood targets
  through the drug pipeline
• Traditional chemical technologies will n to be
  sufficient to identify novel chemical entities
• that can interact with a target- could have
  adverse outcomes during the clinical trial process

Source Company
20
Assuming no increase in technology
Genomics influenced increase in RD Costs
More than doubles From current
2010
3.6
3.6
2
1
2005
3.2
1.6
2
2000
1.6
0.8
1995
Total Annual RD Budget
Annual RD Budget/NCE output
NCEs
21
Assuming moderate increase in technology
Genomics influenced increase in RD Costs
Promise of productivity expansion
2010
2.7
4.4
2
0.6
2005
2.6
1.3
2
2000
1.6
0.8
1995
Total Annual RD Budget
Annual RD Budget/NCE output
NCEs
22
5-10 FROM IMPACT
Most technologies are likely to make an impact
only 5-10 years from today
Value
Delineate disease mechanisms
Mapout biological Pathways
Map out human proteome
Seq Human Genome
Map out human genome
2000 2005 2010
2015
We are still years away from the real impact of
Genomics technology. Most of them have just got
started -Biotech Executive
Integrated technologies include both
experimental and informatics approach
23
5-10 FROM IMPACT
Most technologies are likely to make an impact
only 5-10 years from today
Value
Profile complex diseases
Identify Differential Expression
Identify some Cellular proteins
Identify key Post-translational modifications
2000 2005 2010
2015
It will be a few years before we have a protein
chip that is cheap, fast and accurate -Biote
ch Executive Proteomics will be a big help with
target validation. H however, we still need to
increase speed and improve Productivity -Phar
ma RD executive
Integrated technologies include both
experimental and informatics approach
24
5-10 FROM IMPACT
Most technologies are likely to make an impact
only 5-10 years from today
Assign single Function based On
functional Genomics data
Value
Correlate expression data And protein interaction
data
Correlate gene/protein Expression date with
function
Basic protein Structure Homology queries
2000 2005
2010 2015
Most of the data mining algorithms are pretty
primitive and straightforward today -Biotech
Executive We are facing more explosive data
produced by Genomics technologies. Unfortunately,
the informatics tools are still not there to
allow us to explore them fully -Pharma RD
executive
Integrated technologies include both
experimental and informatics approach
25
THRESHOLD LEVEL OF INVESTMENT NECESSARY
Large investments are necessary to reap the
benefits of technology
LEAD OPTIMIZATION
EXPLORATORY DEVELOPMENT
iNFORMATICS
TARGET VALIDATION

• Threshold annual
• Expenditures
• Key means/technol
• ogies to achieve impact
• at bottleneck

20-40m
20-40m 10-20
20-30m

• Process improvements
• Pharmacogenomics
• Computer aided trial
• design

• Closed loop chemistry
• ADME
• HTS

• Functional Genomics
• tools
• Database subscriptions

• Bioinformatics
• Chemoinformatics
• Clinical Informatics

26
BIOINFORMATICS PRODUCT/SERVICE MODELS
There are three broad organizational models
emerging

• Provide user friendly access to proprietary and
  public gene databases compatible with customer IT
  architecture
• Requires bioinformatic and genomic
  competencies/assets
• Assumes customer does not need to develop
  significant in-house capabilities

• Conduct discrete stages of discovery process
• Requires broad informatic and drug discovery
  capabilities
• Value proposition built on superior informatic
  capabilities

Gene Database Designer
Discovery Services Provider
IT Architects

• Provide off-the-shelf/bespoke informatic
  solutions
• Requires leading edge bioinformatic capabilities
• Assumes customer has in-house skills and
  competencies to be able to leverage and
  manipulate genetic data

The trouble is that bioinformatics is so new,
and the market so ill-defined, that companies are
having difficulty settling on the business model
they will follow In Vivo
Source press search
27
THESE DEMANDS FOR BIOINFORMATICS ARE ADDRESSED BY
THREE MAJOR SERVICE MODELS...
INTEGRATED

• Provide user friendly interface that can access
  both off-the-shelf bioinformatics software and
  more sophisticated IT solutions
• Require extensive IT capabilities

DATABASE FOCUS

• Provides cutting-edge computational solutions to
  discrete components of the discovery process
• Requires extensive expertise in drug discovery
  and bioinformatics capabilities

• Provide access to proprietary and public
  databases, e.g., gene and protein sequences
• Require data acquisition assets (e.g., Genomics
  heritage) along with solid bioinformatics
  capabilities

NARROW
SIMPLE
COMPLEX
ANALYTICAL CAPABILITIES
Source analysis
28
...AND MANY PLAYERS HAVE ADOPTED EACH SERVICE
MODEL
INTEGRATED
RESEARCH ENTERPRISE ASPs
eBioinformatics
DoubleTwist
NetGenics
Viaken
Base4
Bioreason
DATABASE FOCUS
Strand
Celera Genomics
Structural GenomiX
Incyte
Compugen
Hyseq
Tripos
Spotfire
Molecular Simulations
NARROW
DATABASE PROVIDERS
DISCOVERY SOFTWARE
SIMPLE
COMPLEX
ANALYTICAL CAPABILITIES
Source analysis company websites
29
CORE BELIEFS AND CHALLENGES FOR EACH BUSINESS
MODEL
It is not yet clear which if any of the current
approaches will prove sustainable

• Service model
• Gene Database Designer
• IT Architects
• Discovery Services Provider

• Core beliefs
• Databases sufficiently fragmented thus rendering
  inefficient for pharmacos to go it alone
• Ability to remain ahead of pharmacos vis-a-vis
  technological innovation
• Genomic heritage a prerequisite for success
• IT skills are the defining basis of competition
  not knowledge of Genomics
• IT solution will not emerge from existing
  pharmaco IT suppliers
• Ability to remain ahead of other entities
  vis-a-vis technological innovation
• Pharmacos will increasingly seek
  discovery-oriented solutions requiring broader
  skill set (increasing proportion of research
  investments are external)
• Value creating in longer term as provides a base
  for full integration

• Issues
• Multiple public databases challenging role of
  proprietary databases
• Pharmacos are developing skills to create bespoke
  databases in-house
• Real risk that skill could become a commodity
  (e.g., cost of sequencing a bacterial genome fell
  from 12m to 0.5m in 1997)
• Unclear who are the natural owners/developers
  (several pharmacos have thought about this
  longer than we have . . . we need to stay on the
  cutting edge VP SM Molecular Applications
  Group)
• Clear potential for non pharma IT players to
  enter market
• Potential commoditisation of services
• Not clear under which conditions pharmacos will
  outsource discovery functions
• Issues of skills, critical mass and focus present
  real challenges to companies developing from a
  Genomics/IT heritage

Source Team interviews articles
30
CATEGORISING TODAYS BIOINFORMATICS COMPANIES
The traditional genomic companies are polarising
into two categories those that design databases,
and those are broadening their value proposition
to encompass discovery offerings. The new breed
of bioinformatic companies are establishing
themselves in a third category IT architects
Product services providers

• Gene database designers
• Building and distributing annotated gene
  databases and services from public and private
• Alphagene
• Digital Gene Technologies Inc.
• Genome Therapeutics Corp.
• human Genome Sciences Inc.
• Hyseq
• Incyte Pharmaceuticals
• Myriad Genetics
• Sequana Therapeutics

• IT architects
• Building IT systems to enable the sequencing,
  synthesis and access of genomic data
• Base 4 bioinformatics
• Genecodes
• GeneTrace Systems
• Genomica Corp
• Informax Inc.
• MDL Information Systems Inc.
• Molecular Applications Group
• Molecular informatics Inc.
• Netgenics
• Oncormed
• Oxford Molecular
• Pangea Systems Inc.
• PE Applied Biosystems

• Discovery Services Provider
• Conducting discrete stages of the drug discovery
  process using proprietary systems and knowledge
• Acacia Biosciences
• Affymetrix
• Ariad Pharmaceuticals
• Chiroscience (acq. Darwin Molecular)
• Exelixis Pharmaceuticals Inc.
• Genelogic
• Genetech
• Millennium
• Mitokor
• Ontogency
• Pharmagene
• Progenitor
• Structural Bioformatics Inc.
• Xenometrix

Source Annual reports text lines interviews
team analysis
31
MOST OF THE LATEST RD TECHNOLOGIES WERE
DEVELOPED OUTSIDE BIG PHARMA
Chem-informatics
Genomics
Bio- informatics
Transgenic animals
High throughput screening
Pharmaco- Genomics
Combinatorial Chemistry
Proteomics
Molecular modelling
Antisense
32
HT DNA Sequencing
Technology basics
Competitive landscape
Status and current issues

• Typically, a sample of DNA is amplified using
  PCR with specific fluorescent probes for AGTC
  separated by electrophoresis through automated
  technology and DNA sequence is analyzed.
• For sequencing of both genomic DNA and expressed
  genes (cDNA)
• Supplements DNA mapping and positional or
  functional cloning

• Many players are involved in sequencing the
  genome, contributing to both proprietary and
  public databases
• Public Human Genome Project
• Human Genome Sciences
• Incyte Genomics
• Celera Genomics

• Entire human genome will be sequenced by end of
  2001 (Celera appears to be leading the way)
• All 3 billion nucleotides, on 23 pairs of
  chromosomes, composing about 100,000 genes!
• Sequencing does not provide any insights about
  gene function, merely a blueprint for proteins
• Viability of business model for companies only
  sequencing DNA is questionable. Most recognize
  need to move towards functional Genomics and
  protein studies
• Patents on genes or gene fragments (expressed
  sequence tags, or ESTs), without annotated
  function data, are not likely to be approved

DNA SEQUENCING TECHNOLOGY Nucleotides/day
Old method
1,000,000s
1000s
1990
2000
PCR refers to Polymerase Chain Reaction, a
technique for amplifying specific sequences of
DNA Celeras shotgun approach and powerful
computers can sequence 11,000,000 nucleotides per
day
33
HT Proteomics
Technology basics
Competitive landscape
Status and current issues

• Analysis of proteins and protein expression in
  diseased and normal states
• Proteomics deals with two areas
• protein sequence, expression, and modification
  analysis using techniques of protein separation,
  including 2-dimensional electrophoresis (2-DE)
  and protein chips, and identification, typically
  involving mass spectrometry
• 3D structure analysis by X-ray crystallography
  and nuclear magnetic resonance (NMR), as well as
  complex computer modeling. These structures are
  useful for structure-based drug design.

• Fewer companies are engaged in HT proteomics work
  than HT DNA sequencing
• Key players in HT proteomics
• Oxford GlycoSciences
• Large Scale Proteomics Corp.
• Proteome Inc.
• Ciphergen Biosystems
• Players in 3D protein folding (mostly software)
  include
• Structural GenomiX, Inc
• Structural Bioinformatics
• Bio-IT Ltd.

• Protein function depends on 3-D structure and at
  present, even the best computer software is not
  good at modeling protein structure
• Understanding how proteins are modified after
  expression, especially in the presence of drugs
  and/or disease, will dramatically aid drug
  development

PROTEIN ANALYSIS TECHNOLOGY Proteins analyzed/day
100,000s
1000s
lt1
2000
Prototypes
1990
Prototypes, which should be commercial within 2
years, involve high throughput separation
techniques (HPLC) and advanced mass spectroscopy
(MALDI-TOF) Source Science journals, popular
press, public biotechnology reports
34
Biochip Microarrays
Technology basics
Competitive landscape
Status and current issues

• Current market for biochips is about 175
  Million, and is dominated by Affymetrix however,
  many new players are entering the market with
  alternative chip technologies
• Nanogen (electroactive chips)
• Illumina (fiber optic bead-based)
• Sequenom (industrial Genomics with mass
  spectroscopy)
• Ciphergen (protein chips)
• Affymetrix business model It nearly gives
  away a detection machine (175,000) and then
  hopes to make money from the sale of its
  disposable GeneChips (Razor blade approach)

• Biochip microarrays are ordered sets of known
  molecules (DNA, proteins, etc) attached to a
  solid support (silica, fibers, etc) that allow
  for a vast number of parallel experiments in
  miniature.
• DNA chips are made by either building short
  sequences of DNA on chips or by attached pre-made
  oligonucleotides (short pieces of DNA) to the
  chip
• Expressed cDNA prepared from samples is then
  allowed to interact with the DNA on the chips and
  these interactions are detected.
• This same principle can be applied with proteins
  and small molecules

• As of today, chips with 250,000 probes are
  commercially available in near future, probes
  representing entire genomes should be available
• The use of DNA arrays to interrogate biological
  information represents a paradigm change that
  will profoundly alter biology and medicine
• Dr. Leroy Hood
• University of Washington
• Uses for biochip microarrays are exploding
• gene sequencing
• polymorphism identification
• genetic testing
• gene expression profiling
• toxicology analysis
• forensics
• immunoassays
• proteomics
• drug screening

Source Literature, BioInsight
35
GENE CHIP MICROARRAYS ARE SMALL GRIDS CONTAINING
PIECES OF DNA
Technology Basics

• Gene (or DNA) chips are grids
• Each square (feature) on the grid contains the
  same known repeating DNA sequence
• Different squares contain different sequences

T T A
T T C
T T G
T T T
T A A
T A C
T A G
T A T
T G A
T G C
T G G
T G T
T C A
T C C
T C G
T C T
A T A
A T C
A T G
A G A
A C A
A C C
A C G
A C T
A G C
A G G
A G T
A A A
A A C
A A G
A A T
A G A
G T A
G T C
G T G
G T T
G G A
G G C
G G G
G G T
G G A
G G C
G G G
G G T
G C A
G C C
G C G
G C T
Because the DNA sequence is known at each
location on the DNA chip, unknown probe sequences
can be determined by monitoring where on the DNA
chip these probes stick
C T C
C T G
C G A
C C A
C C C
C C G
C C T
C T A
C G C
C G G
C G T
C A A
C A C
C A G
C A T
C G A
Add mixture of unknown flourescently labeled
probes to DNA chip

• Probes stick (hybridize) to squares that have a
  similar sequence to the probe
• A laser reads out which squares the probes stick
  to
• Software makes the information intelligible

36
Approach
BIOCHIPS CAN BE USED IN GENE EXPRESSION
MONITORING AS A POWERFUL TOOL FOR IDENTIFYING KEY
GENES INVOLVED IN OR AFFECTED BY DISEASE PROCESSES
Technology basics

• Compare readouts from chips exposed to healthy
  and diseased samples (probes)
• Differences (dashed boxes) indicate genes that
  may be involved in the disease process
• Gene products (proteins) from these genes may
  serve as good disease targets, therapeutics, or
  markers

DNA
RNA
Healthy
Tissue
Cell
DNA chip
Probes
DNA
RNA
Diseased
Tissue
Cell
Probes
DNA chip
Find healthy and diseased individuals
Isolate healthy and diseased tissues
Isolate RNA from each sample (RNA tells us which
genes are turned on)
Make fluorescently labeled probes from RNA
(probes are pieces of DNA that represent genes
which are turned on)
Expose DNA chips (which have thou-sands of known
genes on them) to probes probes will only stick
to DNA chips in certain loca-tions (see next page)
37
THE COMPETITIVE LANDSCAPE FOR BIOCHIP ARRAYS IS
HEATING UP AS THE TECHNOLOGY RAPIDLY EVOLVES
Competitive Landscape Five example companies and
their technologies

• Uses of biochip
• microarrays continues to expolode
• Gene sequencing
• Polymorphism identification
• Genetic testing (genotyping)
• Gene expression profiling
• Toxicology analysis
• Forensics
• Immunoassays
• Proteomics
• Drug screening
• Many others

• Disposable GeneChip array has oligos attached to
  it by photolithography
• Early leader in biochip development
• Oligos are bound by fluorescent probes

Affymetrix
Nanogen

• Pre-made oligos are bound to reuseable
  semiconductor chip
• Electroactive spots on chip direct and move
  attached oligos, which interact with fluorescent
  probes

Illumina

• Oligos (or drugs, proteins) are attached to
  micro-beads, which self-assemble onto the tips of
  fibers in an optical fiber bundled microarray
• Analyzed by fluorescence with fiber optics

Sequenom

• MassArray chips have oligos attached to them
• Analysis by laser-ionization and mass
  spectroscopy
• Called industrial Genomics

Over 75 public and private Biotech firms make
biochip technology
CipherGen

• ProteinChip array has defined proteins (like
  antibodies) bound to it which interact with
  ligands in the sample
• Analyzed by laser-ionization and mass spectroscopy

Oligos are oligonucleotides, or short (25
bases) sequences of DNA Sources Press reviews,
scientific journals, company reports
38
WHILE AFFYMETRIX HAS DOMINATED THE BIOCHIP
MARKETPLACE, STRONG COMPETITION FROM NEW BIOCHIP
TECHNOLOGIES WILL LIKELY FRAGMENT THE SECTOR
FURTHER
Competitive Landscape
Market Share percent, 1999
1999 Market 176 Million

• Trends in competitive landscape
• Biochip market expected to grow to 1 Billion by
  2005
• New biochip technology players will cut into
  Affymetrixs marketshare
• Use of homemade chips will likely decrease as
  complexity and versatility of commercial chips
  increases
• The market for hardware and bioinformatic
  software for chip detection and data collection/
  analysis will also explode

Source lLiterature BioInsights
39
Pharmacogenomics
Technology basics
Competitive landscape
Status and current issues

• Every individual has a distinct set of
  polymorphism or gene variants. These variants
  could lead to enhanced or diminished responses to
  therapy.
• It applies genetic testing techniques to identify
  these variants that are predictive of a patients
  response to a therapeutic agent
• Pharmacogenomics can be used to
• increase the likelihood of a drugs success in
  the clinic by identifying patients who are more
  likely to have responses to drugs
• rescue previous drugs who failed or were taken
  off the market for safety concerns by identifying
  safe patient populations

• Key players include
• Genset is working on a map of SNPs for clinical
  testing (with Abbott Labs)
• Others companies include
• Affymetrix
• Celera
• GeneLogic
• Incyte
• LJL Biosystems
• Lynx Therapeutics
• Millennium Predictive Medicine

• Pharma community is in agreement that
  pharmaco-Genomics is important - but its effects
  are uncertain
• The FDA has asked us (senior pharma people) to
  come in and discuss pharmacogenomic testing with
  them
• B. Michael Silber
• Director of Clinical Diagnostics
• Pfizer
• Rescuing drugs has the potential to absolutely
  take off, or it might not
• Greg Miller,
• Head of Molecular Profiling,
• Genzyme

40
Lab Automation
Technology basics
Competitive landscape
Status and current issues

• Key players include
• Robotics LJL Biosystems, Robocon, Zymark
• Microplate Perkin Elmer, Molecular Devices,
  Dynex
• Liquid Beckman Coulter, Gilson
• Software Oxford Molecular Group, Tripos, MSI,
  MDL Information systems

• With the explosion of compounds from
  combinatorial chemistry and the accelerated
  identification of gene targets from Genomics, the
  ability to analyze and screen compounds becomes
  critical rate-limiting step. So highly automated
  lab technologies have developed in four major
  areas
• Microplate readers and equipment
• Liquid handling, manipulating, and dispensing
  devices
• Robotics
• Software to control the process

• Likely to see high growth in the next few years
  as lab automation increases
• Miniaturization will lead to lower reagent costs
  likely value shift to equipment and software
• Huge need for quality bioinformatics software
  that is capable of data acquisition/ collection
  as well as data analysis and storage.

Market breakdown by sector 1998, Total market
1.1 Billion
13 CAGR
Source Literature, Genetic Engineering News
41
Database suppliers/designers
Technology basics
Competitive landscape
Status and current issues

• Provides remote access to their proprietary
  database, as well as public ones typically using
  an internet or intranet platform
• Data acquisition skills (e.g., DNA sequencing
  heritage) is a prerequisite for success in this
  segment
• Generally, three main revenue models
• Subscription-based access
• Royalties-based and shared risk
• Fee-for-service

• Key players include
• Celera Genomics (subscriptions to gene database,
  ESTs)
• Incyte Genomics (online Incyte 2.0 LifeSeq
  and LifeExpress databases)
• Human Genome Sciences (exclusive databases for
  Human Gene Consortium)
• GeneLogic (Expression databases)
• AlphaGene (DNA)
• Hyseq (GeneSolutions.com provides access to
  proprietary data)
• Myriad Genetics (ProNet, a proteinprotein
  interaction database)
• Sequana
• Genset (SNPs database)
• Orchid Biosciences (SNPs)
• Oxford GlycoSciences (LifeExpress with Incyte)

• Multiple public databases, like GenBank, are
  challenging the role and importance of
  proprietary databases in many areas (especially
  Genomics).
• Many pharmacos/biotechs are developing their own
  bioinformatics skills to handle databases
  in-house
• Large risk that gene data acquisition skills
  could be commodity (and therefore limit value of
  proprietary databases), e.g., cost of sequencing
  a bacterial genome fell from 12m to 0.5m by
  1997
• Belief that current databases are fragmented and
  inefficient - leading many pharmaco/biotech firms
  to outsource database management

Source Literature press releases
42
Discovery Software Providers
Technology basics
Competitive landscape
Status and current issues

• Provides cutting-edge informatics solutions to
  discrete components of the discovery process,
  e.g., protein folding or CC library selection and
  screening
• Drug discovery process is increasingly seeking
  more sophisticated IT solutions/software that
  require a specialized skill set
• Requires deep expertise in drug discovery as well
  as leading edge bioinformatics/IT capabilities
• Simply put, these are drug discovery tool kit
  companies

• Key players include
• Structural Bioinformatics, Inc. (structure-based
  target id using sophisticated protein structure
  modeling and database)
• Tripos (offers several discovery tools, including
  FlexX, a virtual CC library software)
• Molecular Simulations, Inc. (Pharmacopeia
  subsidiary, software simulates molecular
  interactions of drugs, proteins)
• Compugens LabOnWeb.com (aimed at early gene
  sequence PCR work)
• Bioreason (chemical entity analysis programs)
• Spotfire (decision analytic software aimed at
  researcher productivity)
• Molecular Mining Corp.

• Not clear which activities will be outsourced and
  which will be developed in-house
• Critical mass, skills, and focus are important
  issues for firms developing from a data
  acquisition heritage.
• Value proposition must include superior IT tools.
  

Source Literature press releases
43
Research Enterprise ASPs
Technology basics
Competitive landscape
Status and current issues

• Offer ASP platforms that integrate broad
  databases and sophisticated IT applications,
  coupled with research portal functionality.
• Provides user-friendly interface that offers a
  suite of off-the-shelf bioinformatics solutions
  enabling users to access broad range of
  applications for data and analysis
• Requires leading edge IT capabilities, but does
  not rely on any specific drug discovery or data
  acquisition knowledge.

• Key players include
• DoubleTwist (leader in the research enterprise
  ASP space formerly Pangea)
• eBioinformatics
• Base4 (collaborative knowledge and project
  management platform with database handling
  applications)
• NetGenics (subscription ASP distributing
  computing platform with broad discovery
  applications)
• Genomica (Discovery Manager software suite)
• Viaken (a premier life science ASP for database
  hosting and analytic software)

• Clear potential entry point for non-life science
  IT players
• Potential threat of commoditization of services
• Unclear who are the natural owners of this space

Source Literature press releases