intronIC 2.0.3

Intron classification tool for identifying U2-type and U12-type introns using SVM

intronIC - (intron Interrogator and Classifier)

Version 2.0 - Refactored Edition with Corrected ML Architecture

intronIC is a bioinformatics tool for extracting and classifying intron sequences as U12-type (minor) or U2-type (major) using a support vector machine trained on position-weight matrix scores.

---

Quick Start

Installation

pip install intronIC

Basic Usage

# Classify introns (default model loaded automatically)
intronIC -g genome.fa.gz -a annotation.gff3.gz -n species_name -p 8

# Extract sequences only (no classification)
intronIC extract -g genome.fa.gz -a annotation.gff3.gz -n species_name -p 8

# Train a custom model (optional - most users don't need this)
intronIC train -n my_model -p 8

Test Run

# Download test data or use the included test files
intronIC -g test_data/Homo_sapiens.Chr19.Ensembl_91.fa.gz \
         -a test_data/Homo_sapiens.Chr19.Ensembl_91.gff3.gz \
         -n homo_sapiens_chr19 -p 4

Expected: ~29,000 introns extracted, ~30 U12-type introns found (runtime: 1-2 minutes with -p 4).

---

Documentation

For complete documentation, see the intronIC Wiki:

• Quick Start Guide - Installation, dependencies, resource usage
• Overview - Classification approach and scientific background
• Usage Info - Complete CLI reference
• Output Files - File formats and interpretation
• Technical Details - Algorithm and ML architecture
• Example Usage - Common workflows
• About - Background and motivation

---

What's New in Version 2.0

This refactored version maintains 100% algorithmic fidelity and CLI compatibility with the original intronIC while providing a modernized, maintainable codebase:

Key Improvements

• Corrected ML Architecture: Fixed double-scaling issue and train/test mismatch
  • Single scaling step via RobustScaler with centering (removes composition bias)
  • Configurable augmented features (5D standard or custom)
  • Two-stage optimization (C via balanced_accuracy, calibration via log-loss)
• Modular Architecture: Organized into logical packages instead of a single 6,000+-line file
• Enhanced Code Quality: Type hints throughout, immutable data structures, better error handling
• Bug Fixes: Corrected data leakage in z-score normalization, fixed type_id assignment
• Modern Tooling: Support for pixi and uv package managers
• Improved Documentation: Comprehensive wiki and inline documentation

---

Key Features

• SVM-based classification with probability scores (0-100%)
• Default pretrained model loaded automatically - works for virtually all species
• Streaming mode (default) for ~85% memory reduction on large genomes
• Parallel processing for improved performance (-p 8 recommended)
• Fast runtimes: ~6-10 minutes for human genome with default settings
• Comprehensive metadata including phase, position, parent gene/transcript

---

Scientific Background

Most eukaryotic introns (~99.5%) are spliced by the major (U2-type) spliceosome, while a small fraction (~0.5%) are spliced by the minor (U12-type) spliceosome. U12-type introns have:

• Highly conserved TCCTTAAC branch point motif
• Terminal dinucleotides: AT-AC (~25%) or GT-AG (~75%)
• Functional importance and evolutionary conservation

intronIC identifies U12-type introns using:

1. PWM Scoring: Apply position-weight matrices to 5' splice site, branch point, and 3' splice site
2. Normalization: Convert raw scores to z-scores (prevents data leakage)
3. SVM Classification: Linear SVM with balanced class weights outputs probability scores

For detailed algorithm description, see the Technical Details wiki page.

---

Citation

If you use intronIC in your research, please cite:

Devlin C Moyer, Graham E Larue, Courtney E Hershberger, Scott W Roy, Richard A Padgett. Comprehensive database and evolutionary dynamics of U12-type introns. Nucleic Acids Research, Volume 48, Issue 13, 27 July 2020, Pages 7066–7078. https://doi.org/10.1093/nar/gkaa464

---

Support

• Documentation: intronIC Wiki
• Issues: GitHub Issues
• Discussions: GitHub Discussions

---

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

git clone https://github.com/glarue/intronIC.git
cd intronIC
make install    # Set up development environment
make test       # Run tests

---

License

intronIC is released under the GNU General Public License v3.0.