Edit distance | Smith-Waterman local alignment | The CIGAR convention | GPU backend

QC pass: which records actually are cytb? | A common window, honestly chosen | Edit distance recovers the species | smax is a claim about relatedness | Takeaway

The alignment | What the QC already knows | Frame integrity: codon mode vs plain DNA mode | Frameshift refinement | Interoperability | Where this engine lives

What you get | The recipe

1. Off-target candidate scoring (bounded edit distance) | 2. Adapter contamination scan (local alignment) | 3. Read dereplication into haplotypes | Why these three?

What is PU learning? | When to use xplus | Basic workflow | Model assessment

Installation | 1. The quickest path: sobol_design() | 2. What makes sobol_design special? | 3. Going one level deeper: raw points | 4. Incremental generation with sobol_generator() | 5. Reproducibility and parallel optimisation | 6. Advanced: chaining generators for adaptive sampling | 7. Performance notes | Next steps | Acknowledgements | References | See Also

Introduction | Historical Background | Basic Usage | Key Features | 1. Incremental Generation | 2. Skip-Ahead Capability | 3. Batch Generation | Comparison with Random Sampling | Monte Carlo Integration Example | High-Dimensional Applications | Performance Considerations | Advanced Usage: Parallel Processing | Best Practices | References | Conclusion

1. Description of the package | 2. Hypothetical example dataset | 3. Visualize patterns generated by rosario() function | 4. Assemblage-wide overlap | 5. Null Model Test | 6. Visualizing the null distribution | 7. Conclusion

1. Overview | 2. Data Loading | 3. Organizing detections into time bins | 3.1. Helper function to bin detections for one species | 3.2. Create binned rows for each cervid species | 3.3. Combine species rows into the rosario input matrix | 3.4. Optional: rescale counts to proportions | 4. Running rosario on the real-world dataset | 4.1. Create cyclic shifts and mirror images | 4.2. Visualize patterns generated by rosario() function | 4.3. Assemblage-wide temporal overlap | 4.4. Null Model Test | 4.5. Visualizing the null distribution | 5. Conclusion