1
(No Transcript)
2
Analyze/StripMiner

• Analyze/StripMiner Overview
• To obtain an idiots guide type analyze gt
  readme.txt
• Standard Analyze Scripts
• Predicting on Blind Data
• PLS (Please Listen to Svante Wold)
• LOO, BOO and n-Fold Cross-Validation Error
  Measures
• Albumin Data Set and Feature Selection
• Bio-Informatics

3
Analyze/StripMiner

• Data Processing
• Interface with RECON
• Different Scaling Modes
• Outlier detection/data cleansing

• Modeling
• ANN (Neural Networks)
• SVM (Support Vector Machines)
• PLS (Partial-Least Squares)
• GA-based regression clustering
• PCA regression
• Local Learning
• Outlier Detection (GAMOL)

• Visualization
• Correlation Plots
• 2-D Sensitivity Plots
• Outlier Visualization Plots
• Different Scaling Options
• Cluster Ranking Plots
• Standard ROC curves
• Continuous ROC curves

• Learning Modes
• Bootstrapping
• Bagging
• Boosting
• Leave-one-out cross-validation

• Code Specifics
• Tight Classic C-code (lt 15000 lines)
• Script-Based Shell Program
• Runs on all Platforms
• Ultra Fast
• Use TransScan GE - KODAK
• Doppler broadening
• Macro-Economics Analysis

• Feature Selection
• Sensitivity Analysis
• Genetic Algorithms
• Correlation GA (GAFEAT)
• Method specific

DDASSL
4
Analyze/StripMiner Coding Philosophy

• Standard C code that compiles on all platforms
• WINDOWS and Linux platforms
• Supporting visualizations use Java and/or
  gnuplot
• Flexible GUI with sample problems and demos
• Fastest code possible with efficient memory
  requirements
• Long history of code use with variety of users
  for troubleshooting
• Flexible code based on scripts and operators
• Operates on a numeric standard data mining
  format file

5
Practical Tips for PCA

• NIPALS algorithm assumes the features are zero
  centered
• It is standard practice to do a Mahalanobis
  scaling of the data
• PCA regression does not consider the response
  data
• The ts are called the scores
• It is common practice to drop 4 sigma outlier
  features
• (if there are many features)

6
StripMiner Script Examples

• PCA visualization (pca.bat)
• Pharma-plot (pharma.bat)
• Prediction for iris with PCA (iris.bat)
• Bootstrap prediction for iris (iris_boo.bat)
• Predicting with an external test set example
  (iris_ext.bat))
• PLS and ROC curve for iris problem (roc.bat)
• Leave-One-Out PLS for HIV (loo_hiv.bat)
• Feature selection for HIV (prune.bat)
• Starplots (star.bat)

7
File Flow for PCA.bat Script
num_eg.txt
stats.txt la_sscala.txt iris.txt.txt.txt.txt

• num_eg.txt contains the number of PCAs (2-10)
• usually data are first Mahalanobis scaled
  (option -3 PLS scaling, data only)

8
File Flow for pharma.bat script
num_eg.txt
stats.txt la_sscala.txt dmatrix.txt a.txt
pharmaplot
num_eg.txt has to contain a 4 for a
pharmaplot use pharmaplot.m for visualization
in MATLAB adjust color setting threshold in
pharmaplot.m
9
File Flow For iris.bat Script Predicting Class
stats.txt la_sscala.txt a.txt cmatrix.txt dmatrix.
txt resultss.xxx resultss.ttt results.xxx results.
ttt
num_eg.txt

• For the random seed in splitting routine dont
  use 0 (preserves order)
• The test set is really only for validation
  purposes (answer is known)
• Note descaling from PLS uses la_sscala.txt file
• Notice q2, Q2, and RSME error measures

10
File Flow for iris_boo.bat Script Bootstrap
Validation for Estimating Prediction Confidence
stats.txt la_sscala.txt a.txt resultss.xxx results
s.ttt results.ttt
num_eg.txt

• We use bootstrap cross-validation (e.g., leave 7
  out 100 times)
• Use MATLAB script dos_mbotw results.ttt to
  display results for test set
• Use MATLAB script dos_mbotw resultss.xxx to
  display results training set
• Notice q2, Q2, and RSME error measures

11
Error Measure Criteria
For training set we use - RMSE root mean
square error for training set - r2
correlation coefficient for training set -
R2 PRESS R2 For validation/test set we use
- RMSE reast mean square error for validation
set - q2 1 rtest2 - Q2 PRESS/SD
12
Script for Scaling with an External Test Set

• 3305 scatterplot (Java)
• -3305 scatterplot gnuplot
• 3313 errorplot (Java)
• -3313 errorplot (gnuplot)

13
(No Transcript)
14
Docking Ligands is a Nonlinear Problem
DDASSL
Drug Design and Semi-Supervised Learning
15
Feature Selection (data strip mining)
PLS, K-PLS, SVM, ANN
Fuzzy Expert System Rules
GA or Sensitivity Analysis to select descriptors
16

• Binding affinities to human serum
• albumin (HSA) log Khsa
• Gonzalo Colmenarejo, GalaxoSmithKline
• J. Med. Chem. 2001, 44, 4370-4378
• 95 molecules, 250-1500 descriptors
• 84 training, 10 testing (1 left out)
• 551 Wavelet PEST MOE descriptors
• Widely different compounds
• Acknowledgements Sean Ekins (Concurrent)
• 
  N. Sukumar (Rensselaer)

17
Script for ALBUMIN_LOO.BAT Pls-loo Validation
For Albumin Data
cmatrix.ori dmatrix.ori num_eg.txt
stats.txt la_sscala.txt a.txt results.xxx results.
ttt sel_lbls.txt bbmatrixx.txt bbmatrixxx.txt

• PLS-LOO stands for leave-one-out PLS
  cross-validation
• Training set is in cmatrix.ori and external
  validation set in dmatrix.ori
• External validation set has 999 or 0 in the
  activity field
• Note that we create generic labels and and that
  there is a test set
• Notice the dropping of non-changing features and
  4-sigma ouliers
• Notice the acrobatics for displaying metrics
  (visualize with dos_mbotw)

18
(No Transcript)
19
(No Transcript)
20
PLS Feature Selection Script For Albumin Data
aa.pat bbmatrixx.txt sel_lbls.txt
select.txt
sel_lbls.txt aa.pat aa.tes bbmatrixx.txt bbmatrixx
x.txt

• Do several iterative prunings, typically leave 7
  out 100 x
• Use different seeds
• Number of selected feature example 400, 300,
  200, 150, 120, 100, 80, 60, 50, 45,

21
DDASSL
22
STARPLOT.BAT Starplot for Selected Features for
Albumin
sel_lbls.txt aa.pat
bbmatrixxx.txt
sel_lbls.txt starplot.txt
starplot

• First generate bbmatrixxx.txt which contains all
  sensitivities for (e.g.) 30 boostraps
• using PLS bootstrap option 33
• Generate starplot.txt from bbmatrixxx.txt using
  option 3320
• Use the MATLAB routine starplot.m (operates on
  starplot.txt and sel_lbls.txt)

23
(No Transcript)