Copy Number Variant Optimization

Objective

The objective of this script is to optimize the selection of amplicons (from ampseq data) for the calculation of copy number variants (CNVs) in Plasmodium falciparum through the use of a Generalized Additive Model (GAM). To achieve this, a genetic algorithm (GA) is employed to systematically identify the most informative set of amplicons from a larger pool.

Functionality Overview

1. Input Data:

   • Loci of Interest for Reference: Load reference loci data from an RDS file.
   • Amplicon Coverage Data: Provide the path to the amplicon coverage file.

2. Data Formatting:

   • Use the formating_ampCov function to format the amplicon coverage data, filtering out irrelevant controls and creating additional columns for analysis.

3. CNV Estimation:

   • Employ the estCNV function to estimate copy number variants for each sample. The GAM model takes into account amplicon length and specified loci of interest.

4. Genetic Algorithm:

   • Utilize a genetic algorithm (GA) to optimize the selection of amplicons for each sample. The GA dynamically evolves a population of potential solutions to maximize the accuracy of CNV predictions known CNV fold changes in control strains of P. falciparum.

5. Fitness Function:

   • Define a fitness function that evaluates the performance of each potential solution (set of amplicons) based on the expected and observed fold changes of loci of interest.

6. GA Parameters:

   • Set parameters such as population size, generations, mutation probability, and elitism to guide the genetic algorithm's search for optimal solutions.

7. Run GA:

   • Execute the GA with real-time plotting to visualize the evolution of fitness scores over generations.

8. Extract Results:

   • Extract the optimal set of amplicons identified by the genetic algorithm.

9. Fold Change Calculation:

   • Calculate fold changes for each locus of interest using the optimal amplicon set.