LASSO Logistic Regression Analysis

When to Read External Files

Usage

Rscript scripts/main.R \
  --input_file ./expression_matrix.csv \
  --group_file ./groups.csv \
  --case_group case \
  --control_group control \
  --output_dir ./output/ \
  --nfolds 10 \
  --timeout_seconds 1800 \
  --seed 42

Arguments

Input Format

Expression Matrix (input_file)

Features as rows, samples as columns, CSV or TSV format with feature IDs in the first column.

,Sample01,Sample02,Sample03
TSPAN6,1.8479,1.8318,3.8276
TNMD,0.0349,0.0533,1.3889

Group File (group_file)

CSV or TSV with sample IDs and binary-group labels.

sample,group
Sample01,case
Sample02,control
Sample03,case

Optional Feature File (feature)

One feature per line, or pass a comma-separated feature list directly on the CLI.

TNMD
DPM1
SCYL3

Output Files

Workflow

Step 1: Validate Input

WHEN checking validation rules or parsing behavior, READ: scripts/utils.R and scripts/io.R

• Check file existence
• Read expression matrix and group file
• Verify samples match between files
• Ensure both classes are present with at least 2 samples per class

Step 2: Prepare Modeling Matrix

WHEN checking class encoding or feature filtering behavior, READ: scripts/modeling.R

• Encode case_group as 1 and control_group as 0
• Optionally restrict to a user-supplied feature panel
• Transpose expression data to sample-by-feature format

Step 3: Fit LASSO Logistic Regression

WHEN understanding the statistical method or lambda selection, READ: references/algorithm.md

• Train a binomial glmnet model with alpha = 1
• Run cv.glmnet to select the optimal lambda
• Extract coefficients at lambda.min

Step 4: Save Results and Visualizations

WHEN checking output generation or plot behavior, READ: scripts/run_analysis.R and scripts/plotting.R

• Save flat output files directly into output_dir
• Generate cross-validation and coefficient path PDF plots
• Leave plot titles empty by default unless the user provides custom titles

Methods

LASSO Logistic Regression

The model minimizes binomial deviance with an L1 penalty, shrinking weak coefficients to zero and performing embedded feature selection.

Cross-Validation

cv.glmnet evaluates candidate lambda values across nfolds folds and reports lambda.min and lambda.1se.

Examples

Basic Usage

Rscript scripts/main.R \
  -i ./expression_matrix.csv \
  -g ./groups.csv \
  -c case \
  -t control \
  -o ./output

Use a Feature Panel

Rscript scripts/main.R \
  -i ./expression_matrix.csv \
  -g ./groups.csv \
  -c case \
  -t control \
  -f ./genes.txt \
  -o ./output

Custom Folds and Seed

Rscript scripts/main.R \
  -i ./expression_matrix.csv \
  -g ./groups.csv \
  -c case \
  -t control \
  -n 5 \
  --timeout_seconds 900 \
  -s 123 \
  -o ./output

Custom Plot Titles

Rscript scripts/main.R \
  -i ./expression_matrix.csv \
  -g ./groups.csv \
  -c case \
  -t control \
  --cv_title "LASSO Cross-Validation" \
  --path_title "LASSO Coefficient Paths" \
  --timeout_seconds 1200 \
  -o ./output

Error Handling

Common Errors

IF error persists, READ: references/troubleshooting.md

Testing

Test with Sample Data

# Check help
Rscript scripts/main.R --help

# Run with sample data
Rscript scripts/main.R \
  -i tests/data/expression_matrix.csv \
  -g tests/data/groups.csv \
  -c case \
  -t control \
  -f tests/data/genes.csv \
  --timeout_seconds 1800 \
  -o tests/output

Validation Commands

# Check coefficient output
ls -la tests/output/coefficient.csv

# Check plots exist
ls -la tests/output/lasso_lambda_binary_plot.pdf
ls -la tests/output/lasso_var_binary_plot.pdf

Implementation Checklist

• CLI parsing with optparse
• set.seed() for reproducibility
• requireNamespace() dependency checks
• Session info recording
• Timeout control with --timeout_seconds
• Temp file cleanup
• File reading instructions in SKILL.md
• Modular script structure (<150 lines per file)
• Test data provided
• Error handling with SKILL_* codes
• Scripts in scripts/ directory
• References in references/ directory

Last updated: 2026-04-17 | Version: 1.0.0