Title: Tree Based Methods for Analyzing Tissue Microarray Data
1Tree Based Methodsfor Analyzing Tissue
Microarray Data
- Steve Horvath
- Human Genetics and Biostatistics
- University of California, Los Angeles
2Acknowledgements
- Horvath Lab
- Yunda Huang
- Xueli Liu Ph.D.
- Zeke Fang Ph.D.
- Tuyen Hoang
- UCLA Tissue Microarray Core
- David Seligson
- Aarno Palotie
- Clinicians
- Hyung Kim
- Arie Belldegrun
3Contents
- Statistical issues with tissue microarray (TMA)
data - Random forest (RF) predictors
- RF clustering
- Application of RF clustering to TMA data
- Supervised Learning Methods
4Background TMA data
5Description of TMA data
- TMA data are a high-throughput tool in validating
newly-identified biomarker in genome wide
discovery - Basic technique was summarized in Kononen et al.
1998
6Tissue Microarray (TMA) Technology Kononen et al.
Nature Medicine 1998
- Hundreds of tiny (typically 0.6 mm diameter)
cylindrical tissue cores - densely and precisely arrayed into a single
histologic paraffin block. - From this new array block, up to 300 serial
4-8 ?m thick sections may be produced. - Targets for fluorescence in situ hybridization
(FISH) and protein expression by
immunohistochemical studies.
donor block array block slide
7Non-normal and highly correlated
8Several Spots per Pathology Case Several
Scores per Spot
- Each case is usually represented by 4 or more
spots - gt3 malignant lesions, 1 matched normal
- Maximum intensity Max (1 4)
- Percent of cells staining Pos (0 100)
- Percent of cells staining with the
- maximum intensity PosMax (0 100)
- Spots have a spot grade NL,1,2,..
- Indicator of informativeness
9Histogram of tumor marker expression scores POS
and MAX
Percent of Cells Staining(POS)
EpCam
P53
CA9
Maximum Intensity (MAX)
10P53 and Ki67 Max versus Pos
11Characteristics of TMA data
- Non-normal, discrete, strongly correlated
- Mixed variable types
- Pooling (combining) spot measurements across
every patient - between 1 to 10 spots of different grade
- current strategy pools tumor spots and forms
median, mean, minimum or max - Message tumor marker intensity is measured by up
to 12 highly correlated staining scores?
multicollinearity
12Our main tool are random forest predictors
- Unsupervised analysis of TMA data
- RF clustering
- Supervised Analysis
- RF based pre-validation method
13Background random forest predictorsL. Breiman
1999
14Random Forests (RFs)
- RFs are a collection of tree predictors such that
each tree depends on the values of an
independently sampled random vector
15Classification and Regression Trees (CART)
- by
- Leo Breiman,
- UC Berkeley
- Jerry Friedman, Stanford University
- Charles J. Stone,
- UC Berkeley
- Richard Olshen, Stanford University
16An example of CART
- Goal For the patients admitted into ER, to
predict who is at higher risk of heart attack - Training data set
- of subjects 215
- Outcome variable High/Low Risk determined
- 19 noninvasive clinical and lab variables were
used as the predictors
17CART construction
High 17 Low 83
Is BP lt 91?
No
Yes
High 12 Low 88
High 70 Low 30
Is age lt 62.5?
Classified as high risk!
No
Yes
High 2 Low 98
High 23 Low 77
Classified as low risk!
Is ST present?
Yes
No
High 11 Low 89
High 50 Low 50
Classified as low risk!
Classified as high risk!
18CART Construction
- BINARY RECURSIVE PARTITIONING
- Binary split parent node into two child nodes
- Recursive each child node can be treated as
parent node - Partitioning data set is partitioned into
mutually exclusive subsets in each split
19RF Construction
20Prediction by plurality voting
- The forest consists of N trees.
- Class prediction
- Each tree votes for a class the predicted class
C for an observation is the plurality, maxC ?k
fk(x,T) C - Regression random forest
- predicted value is the average prediction
21Clustering with random forest predictors
22Intrinsic Proximity Measure
- Terminal tree nodes contain few observations
- If case i and case j both land in the same
terminal node, increase the proximity between i
and j by 1. - At the end of the run divide by 2 no. of trees.
- Dissimilaritysqrt(1-Proximity)
23Casting an unsupervised problem into a supervised
RF problem
- Key Idea (Breiman 1999)
- Label observed data as class 1
- Generate synthetic observations and label them as
class 2 - Construct a RF predictor to distinguish class 1
from class 2 - Use the resulting dissimilarity measure in
unsupervised analysis
24How to generate synthetic observations
- Synthetic observations are simulated to contain
no clusters - e.g. randomly sampling from the product of
empirical marginal distributions of the input.
25RF clustering
- Compute distance matrix from RF
- distance matrix sqrt(1-proximity matrix)
- Compute the first 23 classical multi-dimensional
scaling coordinates based on the distance matrix - Conduct partitioning around medoid (PAM)
clustering analysis - input parameterno. of clusters k
- use the Euclidean distance between the resulting
scaling points
26Theoretical Study of RF Clustering
Ref Using random forest proximity for
unsupervised learning, BIOKDD-CBGI'03, 7th Joint
Conference on Information Sciences, Cary, North
Carolina.
27Applying Random Forest Clustering to Tissue
Microarray Data--Application to Kidney Cancer
- Tao Shi and Steve Horvath
28Scientific QuestionCan one discover cancer
subtypes based on the protein expression patterns
of tumor markers?
29Why use RF clustering for TMA data?
- no need to transform the often highly skewed
features - based on ranks of features
- natural way of weighing tumor marker
contributions to the dissimilarity - elegant way to deal with missing covariates
- intrinsic proximity matrix handles mixed variable
types well
30Kidney Multi-marker Data
- 366 patients with Renal Cell Carcinoma (RCC)
admitted to UCLA between 1989 and 2000. - Immuno-histological measures of total 8 tumor
markers were obtained from tissue microarrays
constructed from the tumor samples of these
patients.
31MDS plot of clear cell patients
- Labeled and colored
- by their RF cluster
32Interpreting the clusters in terms of survival
Clustering label Non clear Cell patients Clear cell patients
1 0 92
2 20 215
3 30 9
33Hierarchical clustering with Euclidean distance
leads to less satisfactory results
Cluster- ing label Non clear Cell patients Clear cell patients
1 9 (20) 286 (307)
2 41 (30) 30 (9)
RF clustering grouping in red
34Euclidean vs. RF Distance
35Molecular grouping vs. Pathological grouping
Molecular Grouping
Pathological Grouping
1.0
1.0
0.8
0.8
p 9.03e-05
0.6
0.6
p 0.0229
Survival
Survival
0.4
0.4
327 patients in cluster 1 and 2
0.2
0.2
316 non-clear cell patients
39 patients in cluster 3
50 clear cell patients
0.0
0.0
0
2
4
6
8
10
12
0
2
4
6
8
10
12
Time to death (years)
Time to death (years)
Message molecular grouping is superior to
pathological grouping
36Identify irregular patients
Clustering label Non clear Cell patients Clear cell patients
1 0 92
2 20 215
3 30 9
1.0
0.8
0.6
Survival
0.4
p 0.00522
0.2
50 non-clear cell patients
Message molecular grouping can be used to
refine clear cell definition.
9 irregular clear cell patients
307 regular clear cell patients
0.0
0
2
4
6
8
10
12
Time to death (years)
37Detect novel cancer subtypes
- Group clear cell grade 2 patients into two
clusters with significantly different survival.
K-M curves
1.0
0.8
0.6
Survival
p value 0.0125
0.4
0.2
0.0
0
2
4
6
8
10
12
Time to death (years)
38Results TMA clustering
- Clusters reproduce well known clinical subgroups
- Ex global expression differences between clear
cell and non-clear cell patients - RF clustering works better than clustering based
on the Euclidean distance for TMA data - RF clustering allows one to identify outlying
tumor samples. - Can detect previously unknown sub-groups
39Boxplots of tumor marker expression vs. cluster
Message clusters can be explained in terms of
tumor expression values, i..e in terms of
biological pathways.
40Conclusions
- There is a need to develop tailor made data
mining methods for TMA data - Major differences
- highly non-normal data
- Euclidean distance metrics seems to be
sub-optimal for TMA data - tree or forest based methods work well for kidney
and prostate TMA data