Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
SYSTEM AND METHOD FOR MANAGING AND PRESENTING INFORMATION
DERIVED FROM GENE EXPRESSION PROFILING
BACKGROUND OF THE INVENTION
1. Copyright Notice.
Certain portions of this patent document may be subject to copyright
protection.
While the facsimile reproduction by anyone of this patent document, as it
appears in the U.S.
Patent and Trademark Office patent files or records, is permitted, no other
use or
reproduction is permitted, and the copyright owner reserves all copyright
rights whatsoever.
2. Field of the Invention.
The present invention is directed to certain systems and methods for managing
and
presenting information derived from techniques for monitoring differential
expression of
nucleic acid sequences, e.g., gene expression profiling.
3. Description of Background Information.
Gene expression profiling processes are commonly used to represent a cell's
physiological response to a particular compound, treatment, or disease. For
example, a
January 1, 1999 article by Iyer et al., Volume 283, Science, at pages 83-87
(www.sciencema~.or~), discloses the use of a temporal program of gene
expression to
represent a physiological response of human cells to a treatment --
particularly, the response
of fibroblasts to serum. A cDNA microarray was used, representing over 8,600
distinct
human genes. Fibroblasts, cultured from human neonatal foreskin, were placed
in a quiescent
state by depriving the cells of serum for 48 hours. The fibroblasts were then
stimulated by
adding a medium containing 10% FBS, and the microarray was then used to
measure the
levels of 8,613 different mRNA sequences at 12 distinct times. The microarray
was used to
identify those genes (including expressed sequence tags -- ESTs) which were
substantially
repressed or induced and the extent of repression or induction (i.e., fold
change). Five
hundred seventeen genes whose mRNA levels changed in response to the treatment
were
selected, and graphically depicted in accordance with a hierarchy.
From this information, various proteins could be identified, which were
categorized
according to their biological functions. Those biological function categories
identified were
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
signal transduction, intermediate-early transcription factors, other
transcription factors, cell
cycle and proliferation, coagulation and hemostasis, inflammation,
angiogenesis, tissue
remodeling, cytoskeletal reorganization, re-epithelialization, cholesterol
biosynthesis, and an
unidentified role in wound healing.
S Various technologies are available for expressing large numbers of genes. A
small
sample of the available implementations incorporating those technologies
include SAGE
(serial analysis of gene expression), oligo arrays, and cDNA arrays.
Those technologies produce data identifying large numbers of expressed genes,
and
the extent of their repression or induction. To aid in the analysis of these
large sets of data,
biological computational analysis systems are being developed. An approach
typically used
to create control and treatment probes comprising respective arrays is one
used by Iyer et al.,
in which the data from such arrays is presented in the form of a two-
dimensional cluster
image showing the dispersion of gene clusters that are either up or down
regulated (induced
or repressed).
Databases and wall charts have been provided which facilitate the study of
treatment
data. For example, the Boehringer Mannheim biochemical pathways wall chart and
the Cell
Signaling Pathways Chart, distributed by Zymed Laboratories, graphically
illustrate select
metabolic pathways existing in nature, the interrelationships between various
of the
illustrated metabolic pathways (such as connections between the metabolic
pathways, and
branching points of substrate metabolism), and factors controlling the
direction and the speed
of turnover from one point to another within a given metabolic pathway.
There is a need for a system which will better facilitate the analysis of data
obtained
from expression profiling techniques, to more readily identify key metabolic
pathway
information, mechanisms of action, mechanisms of drug inactivation and
clearance, and
potential side effects. Such a system will preferably also provide meaningful
information that
assists with the identification of the physiological affects of certain
treatments and the
biological function associated with the affected metabolic activity.
2
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
4. Definitions
For purposes of clarification, and to assist readers in an understanding of
the present
invention, and the embodiments disclosed herein, a number of terms used herein
are defined
as follows:
Biological function:
an inferred functional classification of a given gene, protein, nucleic acid
sequence, or
pathway. Some examples of biological functions are metabolism, angiogenesis,
signal
transduction, transcription factors, cell cycle control, regulation of
proliferation, coagulation
and hemostasis, inflammation, and apoptosis.
Enzyme:
Protein that catalyzes biochemical reactions.
Protein molecule:
One or several polypetide chains of amino acids.
Expression profiling:
A process by which gene expression techniques are used to measure and compare
levels of certain nucleic acid sequences (e.g., mRNAs, proteins, genes, ESTs)
in a cell-
derived sample in relation to the levels of the same nucleic acid sequences
from a different
sample or from the same sample at a different time.
Gene:
A sequence of nucleotides specifying a particular polypeptide chain.
Metabolic pathway:
Any individual biological reaction involving a substrate and a product caused
by a
reaction, as well as the catalyst of such reaction. Catalysts of reactions in
metabolic
pathways are typically enzymatic. A metabolic pathway also includes any
related series of
such individual reactions.
Mechanism of action:
A causal link between a variant and a response to the variant, for example,
identifying
which specific, or where within an individual, metabolic pathway or biological
function does
a compound or treatment act to produce a given physiological effect. For
example, if blood
pressure is reduced, the mechanisms of action comprise the specific metabolic
pathways and
biological functions are being acted upon or involved with the reduction of
blood pressure.
mRNA (messenger RNA):
3
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338_
An RNA molecule synthesized from a DNA template -- by the enzyme RNA
polymerase. An mRNA functions as a template for the assembly of a polypeptide
chain, a
process known as translation.
Physiological affect:
Some physiological change or response. A physiological affect could be a state
of a
given biological system (activation or deactivation), for example a change in
high blood
pressure.
RNA:
Ribonucleic acid.
RNA Polymerase:
An enzyme that synthesizes RNA by using DNA as a template.
Transcription:
A process by which an RNA molecule is synthesized by the enzyme RNA polymerase
using DNA as a template.
SUMMARY OF THE INVENTION
In view of the above, the present invention, through one or more of its
various aspects
and/or embodiments, is thus presented to accomplish one or more objects and
advantages
such as those noted below.
An object of the present invention is to provide an improved mechanism for
facilitating the display of meaningful information based upon expression
profiling, such
information facilitating the determination of biological functions involved
with treatments,
compounds, or diseases, the identification of metabolic pathways, and the
identification of
mechanisms of action. A further object of the present invention is to provide
a structure for
organizing and displaying information to enable data mining, whereby
expression profile data
is grouped in accordance with certain metabolic pathway characteristics in a
displayed map.
The present invention, therefore, is directed to a system or method, or one or
more
components thereof, for managing and presenting information derived from
differential
expression of genetic information which can be to model a physiological
response of
biological cells. The system comprises an expression profiling subsystem. The
expression
profiling subsystem generates, from control and treatment sets of cell-derived
samples,
respective sets of sequence data representing a direction and a magnitude of
regulation of
4
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338_
each one of a high number of different nucleic acid sequences. Sets of nucleic
acid
sequences are associated with particular regions on a map of metabolic
pathways of the
biological organism being studied. An overview of the map coordinates may be
provided,
and those areas or regions of the map comprising high concentrations of
affected nucleic acid
sequences may be differentiated from other regions of the map, for example, by
having a
different color. Regions of the map with high concentrations of the affected
nucleic acid
sequences may be viewed in further detail, to view the specific metabolic
pathways involved,
and the role the affected nucleic acid sequences play within such metabolic
pathways.
Alternatively or in addition, an overview may be provided of the map which
identifies
specific affected nucleic acid sequences within a given set of metabolic
pathways, such
indications include a first symbol representing a point of inhibition within
the set of
pathways, second symbols representing biological catalyst locations within the
set of
pathways, and third symbols representing locations of end products of the
illustrated set of
metabolic pathways.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention is further described in the detailed description which
follows,
by reference to a noted plurality of drawings, by way of non-limiting
exemplary
embodiments of the present invention, in which like reference numerals
represent similar
parts throughout the several views of the drawings and wherein:
Fig. 1 is a block diagram of a gene expression profiling data analysis system;
Fig. 2 is a flow diagram of a gene expression profiling process;
Fig. 3 is a flow diagram of a process for managing information derived from
gene
expression profiling;
Fig. 4 is an overview representation of a biochemical pathway map, indicating
the
concentrations of affected nucleic acid sequences at certain coordinates
within the map;
Fig. 5 is a more detailed blown-up view of certain cells within a given area
of the
biochemical pathway map;
Fig. 6 shows a given set of biosynthetic pathways affected by feedback
inhibition;
Fig. 7 is a diagram of a database structure in accordance with the illustrated
embodiment;
5
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
Fig. 8 is a flow chart representing a process performed by the client computer
to
match expression profiling data up with mapped metabolic pathways;
Fig. 9 is an example of an overview display of the metabolic pathway map in
which
related repressed and induced biological catalysts, a point of inhibition and
end products are
represented by symbols; and
Fig. 10 is a flow diagram of a process of identifying BCIs within affected
pathways
with a simplified set of symbols on the overview map display.
DETAILED DESCRIPTION OF THE EXEMPLARY EMBODIMENTS
Referring now to the drawings in greater detail, Fig. 1 shows an analysis
system 10
according to the illustrated embodiment of the present invention. An
expression profiling
subsystem 12 is provided, which is coupled to a client computer 14. Client
computer 14
comprises, among other elements, a browser application 16, a human interface
18, and a
display 20. Human interface 18 may comprise any standard or other interface
for facilitating
human interaction with and control of client computer 14, including, for
example, a keyboard
and a mouse. Client computer 14 is coupled to a host computer 24 via a network
connection
illustrated in Fig. 1 as an intranet. Host computer 24 is connected to a
database 26.
Expression profiling system 12 may comprise, for example, an Affymetrix cDNA
array. It generates, from control and treatment sets of cell-derived samples,
respective sets of
sequence data representing a direction and a magnitude of regulation of each
one of a high
number of different nucleic acid sequences.
Client computer 14, together with human interface 18, display 20, and browser
application 16, allows a user to operate analysis system 10. Client computer
14
communicates with database 26 through intranet 22 and host computer 24.
Expression
profiling subsystem 12 obtains the expression profiling data and stores that
data in an
organized fashion on database 26.
Host computer 24 is provided with, among other elements, an analysis
application 27
for carrying out certain analysis process steps associated with expression
profiling and
managing the data acquired from the expression profiling. A database server
software
component 28 is provided for handling and acting on database queries and
responses.
Fig. 2 generally shows an expression profiling process in accordance with the
illustrated embodiment. In an initial step S2, sequences are generated based
upon a baseline
6
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338_
sample (otherwise referred to as a control sample) of cells. One or more
differentiated
sequences may be generated based upon treated samples, i.e., samples of cells
based upon
those cells entering into a diseased state or being treated with a particular
compound. After
performing each of respective steps S2 and S6, a cluster algorithm S4 and S8
is performed, in
which similar sequences, including expressed sequence tags (ESTs) are grouped
together.
Clustering of gene sequence pieces allows redundancies to be eliminated, as a
gene
expression array will typically identify not only full gene sequences or full
mRNA, but will
also identify ESTs, which comprise shorter pieces of the full sequence. The
total number of
sequence pieces within a given cluster may be considered to represent the
total number of
genes repressed or induced having a particular sequence.
An alternative method of clustering is to use the expression data to cluster
by
expression patterns, i.e., similar profiles over a course of time. This
approach would allow
comparison between genes having known functions with genes having unknown
functions to
assist in identifying the unknown functions, such as is done by Iyer et al. in
the above-
identified article.
In order to determine whether the gene clusters have been substantially
affected (i.e.,
either repressed or induced), the number of genes generated in the baseline
sample is
compared with the number of genes generated and clustered in each cluster in
the treated
sample or samples, to produce, for each treated sample, an indication of
whether the gene
cluster was regulated and the extent and direction of that regulation.
More specifically, by way of example, a sample of cells may be sequenced using
an
expression profiling array, such as an Affymetrix GeneChipTM probe array for,
for example,
the human genome, which is capable of detecting over 6,000 sequences for that
genome.
Affymetrix provides a GeneChipTM fluidics station which automates the
hybridization of
nucleic acid targets to a probe array cartridge, and thus controls the
delivery of reagents and
the timing and temperature for hybridization. Each fluidics station can
independently process
four probe arrays at a given time.
Accordingly, each target may be prepared from a set of cell dishes by
isolation of
RNA over a course of time. The treatment of those cells may be emulated by
adding, for
example, serum thereto. At predetermined intervals, a small amount of the
fluid is removed,
and the cells are put in a quiescent state to stop the reaction time.
Accordingly, a large set of
targets, having a predetermined amount of liquid (e.g., .5 ml each) is
produced. The
7
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
GeneChipTM fluidics station will then automatically hybridize each target,
i.e., it will extract
all the RNA and label the RNA by adding a chemical tag to each molecule, and
control the
delivery of the resulting liquid to the probe arrays to facilitate the
obtaining of sequencing
information regarding the mRNAs. This is done by the probe arrays exposing the
target to
light at a predetermined location and measuring the photons collected at
various locations
within the arrays. The amount of mRNA (or an EST) is then ascertained based
upon the
signal strength of the reading given by the probe at the appropriate location
corresponding to
that sequence or sequence segment.
Fig. 3 is a flowchart of an analysis process performed by the illustrated
embodiment.
In a first step S20, gene expression profiling is performed, at which time
respective sets of
sequence data are generated from control and treatment sets of cell-derived
samples/targets,
and the obtained data includes information regarding the direction and
magnitude of
regulation of each one of a high number of different nucleic acid sequence
clusters. Once
gene expression profiling is performed at step 520, a set of data D2 is
produced which
comprises the identified sequences and associated regulation information.
Then, at step 522,
each sequence cluster is matched to a biological catalyst identifier (BCI). In
the illustrated
embodiment, the BCI may comprise, for example, an EC number. EC numbers are
part of a
known system for enzyme classification. Each EC number comprises a first
number which
refers to one of six main subdivisions, a second number which indicates a
subclass, a third
number indicating a sub-subclass, and a fourth number which represents a
serial number.
The major EC classes include (1) oxodoreductases-redox reactions (2)
transferases -- transfer
a group (CH3), (3) hydrolases -- cleavage, H20, (4) lyases -- cleavage by
elimination, (5)
isomerases -- geometric changes, and (6) ligases -- coupled to ATP hydrolysis.
As an
example of some subclasses, the oxoreductases are as follows:
1. Oxoreductases
1.1. CHOH Boners
1.1.1 NAD + or NADP+ acceptor
1.1.2 Cytochrome acceptor
1.1.3. Oxygen acceptor
1.1.5 Quinone acceptor
1.1.99 Other acceptor
8
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338_
At step 524, each cluster of affected sequences, i.e., sequences that have
been
significantly regulated (by at least twofold) is categorized in accordance
with its cluster,
whether it was up or down regulated (i.e., induced or repressed, respectively)
and the extent
of regulation, and further, the number of regulated sequences or sequence
segments (ESTs)
falling within a given cell of contour plot 30 is summed and binned in
association with that
cell. This is performed at step 526.
At step 528, a summed section of a detailed map view is displayed, which
includes
metabolic pathways corresponding to substantially affected sequences. Fig. 4
illustrates a
contour plot view 30 of a biochemical pathway map, which illustrates and helps
clarify the
acts performed in steps S24 and S26 of the process illustrated in Fig. 3.
More specifically, Fig. 4 is a representation of a contour plot view of a
biochemical
pathway map. In the illustrated embodiment, the map corresponds to the
biochemical
pathways wall chart by Boehringer Mannheim. The map may comprise graphic
representations of biochemical pathways which are identical or comparable to
the Boehringer
Mannheim wall chart, or any other appropriate set of graphical representations
of
biochemical pathways, where a given pathway, or point within a pathway, is
associated with
a particular set of coordinates within the map. In the illustrated embodiment,
a matrix of cells
is provided, comprising fourteen columns along the X direction (Xl-X14) and
eight rows
along the Y direction (Yl-Y8). The contour plot view shown in Fig. 4 shows
whether the
number of sequences having an EC number within a given cell is within one of
five
prescribed ranges. Those ranges are depicted by a different pattern, and
include 0,1, 2-3, 4-5,
6-7, and 8-10. By way of example, the cell at coordinates X8, Y7 has one
sequence having
an EC number falling within that cell. The cell at X7, Y7 has seven sequences
with an EC
number falling within that cell. Accordingly, the cell X8, Y7 is illustrated
as falling within
the second range, 1, and the cell at X7, Y7 is shown as having a number of
sequences falling
within the range 6-7.
While patterns are shown in Fig. 4 in order to differentiate between different
ranges of
sequences having an EC number falling within a given cell, it is preferred
that the ranges be
depicted with the use of a coloring scheme. By way of example, the range 1
could be
represented by the color purple, while the range 2-3 is represented by the
color green, the
range 4-5 is represented by the color yellow, the range 6-7 is represented by
the color orange,
and the range 8-10 is represented by the color red.
9
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
The view provided by the contour plot shown in Fig. 4 can thus provide a quick
overall view of the activity throughout the various areas of the pathway map,
and those areas
having yellow, orange and red colors indicate those areas with the most
activity.
Accordingly, one can select areas in accordance with the amount of activity to
view a more
detailed view of the map.
Fig. 5 shows a small portion of a biochemical pathway map which illustrates
various
aspects of certain biochemical pathways at prescribed coordinates x"_l, x" and
xn+~ along the
x direction, and ym_1 , ym, and ym+i along the y direction. The map comprises
graphical
representations of metabolic pathways. Those graphical representations
comprise individual
graphic representations of such items as substrates, products, biological
catalysts (BCs),
inhibitors, biological functions, and pathway directions (including unique
graphical
identifiers showing a direction of a pathway in one direction versus the
opposite direction,
and an amphibolic pathway direction which indicates that the reaction can go
in either
direction).
1 S More specifically, as shown in Fig. 5, a plurality of pathway direction
symbols 40a-
40d are provided in the section of the map shown in Fig. 5. The use of an
arrow at each end
of the illustrated lines 40b and 40c indicates that the pathway direction is
amphibolic. A
plurality of substrate/product symbols 42a-42c are provided which represent
substrate/productsl, 2 and 3~ Those symbols may comprise, for example, text
identifying a
given compound which may serve as either a substrate or a product, depending
upon the
direction of the chemical reaction. Each biological catalyst or set of
biological catalysts
associated with the particular pathway, including biological catalyst(s)~ and
biological
catalyst(s)2 in the illustrated embodiment, is illustrated with a respective
biological catalyst
symbol 44a,b adjacent to the pathway direction symbol.
A block is provided for indicating a biological catalyst symbol 44a and 44b.
These
symbols may simply comprise a textual representation of the common
nomenclature for the
given biological catalyst, which typically will comprise an enzyme in the case
of metabolic
pathways. BCI (biological catalyst index) symbol 46a, 46b is provided adjacent
its respective
biological catalyst symbol 44a, 44b, and in the illustrated embodiment simply
comprises a
numerical representation of the BCI. Any inhibitors will be represented with
inhibitor
symbols 48a, 48b, which, in the illustrated embodiment, may simply comprise
text
representing the inhibitor using standard nomenclature.
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
The biological function with which the metabolic pathways in a certain region
of the
map are associated may be represented with a biological function symbol 50,
which, in the
illustrated embodiment, comprises a text representation of the biological
function using
common nomenclature. Some example biological functions include fatty acid
oxidation,
S carotenoids, and ketone bodies. Other functions include, for example,
sulphur metabolism
and pterine biosynthesis.
In the illustrated embodiment, one or more of the graphic representations may
have a
unique color to identify the type of information it is representing. For
example, the text
serving as BCI symbols 46a, 46b may be in green, the text serving as the
biological catalyst
symbols 44a, 44b may be magenta or aqua, the text serving as the inhibitor
symbols 8a, 48b
may be the color brown, and the text serving as the biological function symbol
50 may be the
color blue. Additional or alternative coloring schemes may be used. Also,
unique graphical
patterns may be used in addition or instead of colors to facilitate the
viewer's ready
identification or classification of a particular symbol as representing one
type of information
versus another. The enzymes shown in Fig. 5 may have two colors, one if it is
induced (up
regulated), and another if it is repressed (down regulated). Accordingly, in
the illustrated
embodiment, biological catalysts 44a and 44b are magenta and aqua,
respectively, indicating
that biological catalyst (s)1 44a was induced, while biological catalyst (s)2
44b was repressed
(down regulated).
By mapping sequences obtained from expression profiling techniques to specific
symbols within a metabolic pathway map, such as shown in Fig. 5, the
information provided
by the expression profiling data can be quickly related to meaningful pieces
of information
relevant to key concerns associated with the treatment, disease, or compound
being applied to
the tested cells. The visualization of the results of the expression profiling
experiment is
enabled by identifying such valuable pieces of information as biological
function
(represented by a biological function symbol 50), metabolic pathway
(represented by a set of
graphical representations forming a given metabolic pathway at specific
coordinates within
the map), and a mechanism of action (the identification of which will be more
fully described
by the use of an example below).
This can have significant benefits in the evaluation of treatments and
compounds, for
example, allowing the identification of mechanisms of action, mechanisms of
drug
inactivation and clearance, and potential side effects.
11
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
Fig. 6 is an illustration of a select group of related pathways. The related
pathways
shown in Fig. 6 may correspond, for example, to a number of identified
biological catalysts
on the map as depicted in the "big picture" view provided in Fig. 9, which
will be described
further below.
Fig. 6 shows a composite pathway comprising a plurality of pathways (pathways -
pathway9). Each illustrated pathway (pathways - pathway9) may comprise one or
more
metabolic pathways, as such pathways exist in nature. In this regard, a
reference may be
made, for example, to the Boehringer Mannheim biochemical pathways wall chart.
The
specific pathway shown in Fig. 6 can be viewed to identify mechanisms of
action, and
toxicology and side effects.
Many biochemical pathways involve a long chain of distinct chemical reactions
catalyzed by distinct enzymes. The first committed step in a biosynthetic
pathway is often
regulated by the final product of the pathway through a process called
feedback inhibition.
Inhibition of a specific enzyme along a metabolic pathway leads to increased
levels of
intermediate chemicals preceding the point of inhibition, and decreased levels
of metabolites
following the point of inhibition.
In the composite pathway shown in Fig. 6, a point of inhibition A is shown.
Enzymes
in the pathway following the point of inhibition A are repressed, while
enzymes in another
direction following the point of inhibition A are induced. When this occurs, a
pathway is
inhibited which prohibits the formation of a given final product, and removes
any feedback
inhibition. Specific enzyme inductions or repressions in response to a disease
state, or
application of a drug to the system, can be used to identify those pathways
which are affected
by the disease or drug.
For example, as shown in Fig. 6, a drug may be found to decrease serum
cholesterol
levels when given to an animal, and that drug may work by an unknown mechanism
which is
revealed by the graphically-represented pathways. Since cholesterol
biosynthesis occurs
primarily in the liver, the liver can be removed and mRNA can be isolated
therefrom. Using
expression profiling techniques, one can determine how this inhibition affects
the mRNA
level of thousands of enzymes acting in dozens of pathways. The pathways whose
enzyme
levels are significantly affected by drug treatment indicate the pathway and
likely suggest a
mechanism of drug action.
12
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
This is the case for inhibitors of hydroxy-methyl-glutaryl-CoenzymeA (HMG-CoA)
reductase, which is the first step in cholesterol biosynthesis. This step is
shown at the top of
Fig. 6.
Along pathways, HMG-CoA is converted to long-chain fatty acids by way of
Acetyl-
CoA in two reaction steps (not specifically shown in detail in Fig. 6). In
another direction,
HMG-CoA is converted to a five carbon isoprenoid via a pathway4, and then to a
ten carbon
geranyl via a pathways. After another pathway, a product 15 carbon farnesyl is
produced.
Another pathways produces a 30 carbon squalene, which is then converted to the
steroid
lanosterol, via pathways. Then, after pathway, which comprises a plurality of
other reaction
steps, cholesterol is produced.
When the drug (HMG-CoA reductase inhibitor) is applied to the liver, and
expression
profiling is performed on the treated liver, the HMG-CoA reductase and enzymes
involved
in fatty acid metabolism (which go along the direction of pathways-pathway3)
are induced,
and the enzymes involved in the formation of cholesterol are repressed.
The identification of pathways of drug metabolism and elimination is done
similarly.
Most drugs are metabolized by oxidation to a more reactive species than
conjugation to a
sugar or other molecule that is recognized in the kidney for elimination. The
oxidative step is
catalyzed by one or more of over 200 enzymes, including cytochrome P 450
enzymes,
followed by conjugation by conjugating enzymes in the liver. These enzymes may
be
induced directly by the drug, or because the drug competes with a normal
substrate, in which
case less of the normal product is produced by the enzyme pathway, and
feedback by that
product is reduced.
Induction of some genes is indicative of toxic effects. A variety of enzymes
involved
in drug metabolism are induced in tumor cells (P450 4 Fl) and the induction by
a drug can
indicate that a drug is potentially tumorigenic. In addition, metabolism of a
drug may create
toxic metabolites, and may induce peroxidation and proteolytic cascades, which
can indicate
that a drug or drug metabolite is causing cell death or damage.
Fig. 7 generally shows, in a block diagram, the structure of the database 26
illustrated
in Fig. 1. Database 26 comprises, among other elements, seven tables as
illustrated in Fig. l,
including tables (an experiment), tablet (data), table3 (sequence), table4
(BCI link), tables
(BCI number), table6 (map link), and table? (coordinate).
13
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
The experiment, tablel, is populated by expression profiling subsystem 12 at
some
point in time. It includes experiment identifiers (ExpID) and associated
experiment names
and experiment conditions. Tablet includes the data obtained from the
experiment, including
the experiment identification (ExpID) the sequence identification, sequence
ID, and the fold-
s change of each sequence that has been identified as being affected. Tablel
is linked with
Tablet by means of the variable ExpID. Tablet holds an associated sequence ID
and fold-
change values in association with each ExpID value. The sequence ID value
within tablet is
associated with a corresponding indexed sequence ID in table3 which serves as
a sequence
table. For each sequence ID, additional variables are associated therewith,
including an
accession variable, and a description of the sequence.
A BCI link table4 is provided which is linked to tablet and table3 in
accordance with
a sequence ID index thereof. BCI link table4 associates with each sequence ID
values
including BCI ID, a sequence/link value, and a link score. Each BCI ID has an
associated
BCI number (BCI) which is listed in tables. Each BCI ID of table4 and of
tables is linked to
a BCI ID index provided in a map link table6. Each BCI ID has a coordinate ID
associated
therewith, which is provided within map link table6. Map link table6 is linked
to coordinate
table? by means of a coordinate ID value. Coordinate table? provides values
associated with
each coordinate ID value, including an x coordinate of the biochemical pathway
map, a y
coordinate of the biochemical pathway map, and a biological function
associated with the
given location on the map per the corresponding x and y coordinates. The
database 26 may
be implemented, in the illustrated embodiment, in accordance with the third
normal form of
relational database. It is noted that most of the actual data is stored in
tablet, table3, tables
and table?, while link tables, table 2, table4 and table6 are provided to
primarily minimize
redundancy in the database.
Linking tables, tablet, table4, and table6, facilitate the many-to-many
relationships.
Such exist between experiments and genes -- many genes are affected in a given
experiment,
and many experiments may be done with each gene. There are also many-to-many
relationships between genes and BCI numbers (e.g., EC numbers). For example, a
multi-
functional gene may have many EC numbers, and many similar genes could have
the same
EC number. Many-to-many relationships also exist between BCI numbers and
mapped
coordinates. For example, if the BCI number comprises an EC number, and the
map
comprises or is modeled after the Boehringer Mannheim biochemical pathways
wall chart,
14
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338-
one EC number can easily appear more than once within a coordinate or in
multiple
coordinates, and each coordinate can have many EC numbers.
Fig. 8 is a flowchart illustrating a process of handling data, which is
performed by
analyzing system 10 in connection with its use of database 26. In a first step
540, experiment
S data is read and stored in tablel. Then, in step 542, the act of storing
sequence data in table3
is performed. The experiment data stored in tablel includes, among other data,
the
experiment ID (ExpID), the experiment name (ExpName), and the conditions of
the
experiment. The sequence data stored in table3 includes the sequence id, the
accession
number corresponding to that sequence, and description data concerning the
sequence. In
step 544, the fold change per sequence (or per sequence cluster) is
determined, and that
information is stored in tablet and related to other data including ExpId and
the sequence ID.
In step 546, the BCIs are linked to sequences. Table4 is then used to link the
sequences to
the BCI data in tables.
In step S48, the BCIs are linked to map coordinates of the map. Link table
table6 is
used to link the BCIs to the coordinate data in Table7.
Fig. 9 shows another overview display of the map. In this view, a point of
inhibition
60 is displayed with a first symbol 60 (which is a square in the illustrated
embodiment) at a
specific location within a particular cell of the map corresponding to the
point in the pathway
at which the inhibition occurs. Second symbols 62a - 621 represent enzymes
which
correspond to sequences affected by the treatment. One color (dark gray in
Fig. 9) is used to
represent enzymes which are induced, while another color (white in Fig. 9)
represents
enzymes which were repressed. Third symbols 64a and 64b represent end products
of the
illustrated pathways. The symbols shown in Fig. 9 are all on a common
composite pathway.
End product symbol 64a is shown as dark gray because it is the end product of
the pathway
corresponding to the induced enzymes, while end product symbol 64b is shown as
white
because it is the end product corresponding to the pathway which is populated
by enzymes
which were repressed.
The analysis application 27 may be configured so that various display modes
are
provided, including a first display mode in which the contour map view is
provided as shown
in Fig. 4, and a second display mode in which respective overview pathways are
provided as
shown in Fig. 9. When in the second mode, each composite pathway may be
separately
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338.
illustrated on its own, or one map may be provided on which the unrelated
composite
pathways are all indicated.
A third display mode may be provided in which a detailed view of the map is
provided. This mode may be entered by the user selectively choosing a detailed
map at any
desired set of coordinates, by simply clicking on the desired coordinates in
an overview
display in either of the first and second display modes.
Fig. 10 is a flow diagram of those steps performed by analysis application 27
to create
the overview display shown in Fig. 9. In a first step 550, the act of
determining specific
coordinates of BCIs is performed. In a next step S52, the BCIs are determined
which are
common to the same pathway. If there is more than one separate unrelated
composite
pathway, a plurality of sets of BCIs are determined and separately
categorized. In step 554,
the induced BCIs of a given common pathway are displayed, with one color
representing
induced BCIs and another color representing repressed BCIs.
In step 556, subcoordinates of the point of inhibition are determined -- if
there is a
point of inhibition, i.e., if one side of the common pathway includes all
repressed BCIs, while
another side of the common pathway includes all induced BCIs. This point is
displayed at
the appropriate location within the biochemical pathway map with a second
symbol.
At step 558, the subcoordinates of the end products of the common composite
pathway are determined, and those points are displayed with a third symbol,
with one color
representing the end product of a pathway portion corresponding to induced
BCIs and
another color representing an end product corresponding to the end of a
portion of a path
corresponding to the repressed BCIs.
The point of inhibition may, for example, be determined by identifying the
point
along a pathway at which the enzymes switch from one affected state (e.g.,
induction) to
another state (e.g., repression). The end products may, for example, be
presumed by
determining the point along the pathway at which the enzymes are no longer
affected, or with
the use of data known about the relevant pathways.
Another, more specific embodiment of the present invention will now be
described.
This embodiment is merely an illustrative example.
Initially, a database is created which relates EC (enzyme commission) numbers
to
coordinates on the Boehringer Mannheim biochemical pathways wall chart. This
database
contains current descriptions for all EC numbers and other information
pertaining to the EC
16
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
numbers. Descriptions of the EC numbers and other enzyme data are publicly
available, and
may be obtained from the website http:/www.expasy.ch/txt/enzyme.get. A
database may
then be created linking the EC numbers with specific map coordinates
corresponding to the
Boehringer Mannheim biochemical pathways wall chart.
Once expression profiling is performed, and experiment data is obtained, EC
numbers
are assigned to the sequence clusters obtained in the experiment. This may
involve a list of
GenBank accession numbers corresponding to those affected genes affected more
than two
fold in a set of profiling experiments. GenBank records are available at
http://www.ncbi.nom.nih.gov/entrez/, and may be parsed for the pattern of
numbers in an EC
number (#.#.#.#). For every occurrence of an EC number in the GenBank file, a
GenBank
accession number and corresponding EC number may be written to a text file for
loading into
a database. The following is a sample GenBank file:
LOCUS 4191746 375 as 27-JAN-1999
DEFINITION alcohol dehydrogenase; ADH.ACCESSION 4191746PID g4191746
DBSOURCE GENBANK: locus L30113, accession L30113KEYWORDS
SOURCE baboon. ORGANISM Papio hamadryas
Eukaryota; Metazoa; Chordata; Vertebrata; Mammalia; Eutheria;
Primates; Catarrhini; Cercopithecidae; Cercopithecinae; Papio.
REFERENCE 1 (residues 1 to 375)
AUTHORS Cheung,B., Holmes,R.S., Easteal,S. and Beacham,LR.
TITLE Evolution of Class I Alcohol Dehydrogenase Genes in Catarrhine
Primates: Gene Conversion, Substitution Rates, and Gene Regulation
JOURNAL Mol. Biol. Evol. 16 (1), 23-36 (1999)
FEATURES Location/Qualifiers source 1..375
/organism="Papio hamadryas"
/db xre~"taxon:9557"
/tissue type="kidney" Protein 1..375
/note="ADH"
/product="alcohol dehydrogenase"
/EC number="1.1.1.1" CDS 1..375
/note="putative"
/coded by="L30113:53..1180"ORIGIN
17
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
1 mstagkvikc kaavlwevkk pfsieeveva ppkahevrik mvavgicrsd dhvvsgtlvt
61 plpailghea agivegvgeg vttvkpgdkv iplftpqcgk crvcknpesn ycfkndlsnp
121 rgtmqdgtrr fteggkpihh flgistfsqy twdenavak idaasplekv cligcgfstg
181 ygpavkvakv tpgstcavfg lggvglsavm gckaagaari iavdinkdkf akakelgate
241 cinpqdykkp iqevlkemtd ggvdfsfevi grldtimasl lccheacgts vivgvppdsq
301 nlsinpvlll tgrtwkgaif ggflcskesvp klvsdfmakk fsldalitnv lpfekinegf
361 dllrsgksir tilmf//
If no EC number is available in the GenBank file, the nucleotide or amino acid
sequence may be obtained from the GenBank file which corresponds to a
particular cluster
obtained from the expression profiling, and a BLAST sequence alignment may be
performed,
which may be performed by accessing the publicly available application through
http://www.ncbi.nlm.nih.gov/cgi-vin/BLAST/nph-newblast?Jform=0. The GenBank
file may
then be fetched for each sequence that aligns with an expect value (E value,
right-most
1 S column in the BLAST results) that is less than 1 e-30, and by looking for
EC numbers in
these related sequence files. If an EC number is present, the accession number
for the gene
affected in the expression profiling experiment can be recorded, and the
expect value from
the sequence alignment may be recorded as well, along with the EC number or
numbers
found in the related sequence file or files.
At this point, the database can be created, as described previously in this
document.
In this regard, in accordance with the specific embodiment now being
described, database 26
as shown in Fig. 1 may comprise an ORACLE database, and host computer may
comprise a
Silicon Graphics Origin 2000 computer. These items are merely illustrative,
and are not
meant to limit the invention in any way. Other computer systems, databases,
and database
structures may be used.
Analysis application 27 may be implemented with use of a Netscape FastTrack
WWW server using standard HTML and Perl. The Perl modules which may be used to
implement this application include (1) DBI/DBD -- a database interface for
communicating
with a remote pathmap database, (2) CGI -- for generating HTML code, (3)
PGPLOT -- an
interface to compiled PGPLOT Fortran libraries for creating contour plots, (4)
GD -- a
graphical drawing module for cropping a GIF image produced by PGPLOT and for
drawing
polygons and rectangles used for background coloring, (5) MLDBM -- a Perl
module that
18
CA 02362544 2001-08-23
WO 00/50889 PCT/US00/04338
allows creation of a persistent mufti-level data structure to implement image
map shape data,
and (6) ImageMagick -- a module for performing image processing, so that the
background
created with GD can be used to create masks, overlays and background coloring.
The application may be configured so that a user can connect to a path map web
page
through the use of browser application 16, select an experiment, and query the
database to
select the wall chart coordinates of genes affected more than two-fold in the
experiment. The
number of genes mapped to each map coordinate are binned, and a contour plot
of hits per
coordinate may be displayed, for example, as shown in Fig. 4. Other displays
may be
provided, as well, such as those shown in Fig. 9. The user may move the cursor
with the use
of the mouse to the position on the map image to see the biological function
corresponding to
that area of the map, and can click on that particular cell of the map to
obtain a more detailed
view of the pathway information, such as that shown in Fig. 5. In this regard,
if the
Boehringer Mannheim biochemical pathways wall chart structure is used, it is
modified to
illustrate the induced and repressed genes, as well as the EC numbers in
association with the
identified enzymes corresponding to those genes. The enzymes corresponding to
affected
genes are colored based upon whether the gene was repressed or induced.
Specifically, the
enzyme may be represented with magenta text if the corresponding gene cluster
was induced,
cyan if it was repressed, and green if two or more gene clusters with the same
EC number
were affected in opposite directions. The interface provided to the user
through browser
application 16 is displayed on display 20, and may provide a mechanism for
allowing the user
to click on the accession number in order to obtain information on a
particular gene and all
available experiments pertinent to the gene. A mechanism may also be provided
to allow
clicking on a particular EC number to obtain all information relating to that
EC number. In
addition, the analysis system 10 may be provided with a search tool to allow
the user to
submit queries by any given parameter to obtain information related to that
parameter. For
example, the user may query by accession number or gene description to find
information for
a specific gene of interest.
19
