# Install CRAN packages
install.packages(c("Rcpp", "ggplot2", "dplyr", "stringr", "devtools"))

# From GitHub (replace 'username' with actual repo owner)
library(devtools)
install_github("limeng12/multiplexILP")

library(multiplexILP)

# Generate simulated data
primer_single_data <- generate_single_primer_data(
  n_primers = 150,
  n_loci = 10,
  num_channels = 2,
  size_gap = 10,
  seed = 123
)
View( as.data.frame(primer_single_data[1:9]))

# Generate ILP model
ilp_result <- gen_ilp_single_primer(primer_single_data)


# Solve with SCIP (ensure SCIP is in PATH)
unlink("sol.txt")
system(paste0("scip -f ", ilp_result$lp_file, " -l sol.txt"))

# Parse results
parsed_results <- parse_scip_single_primer_solution(
  scip_output_file = "sol.txt",
  primers_info_file = ilp_result$primers_info_file,
  output_prefix = "single_primer_analysis"
)

# Visualize
plots <- generate_simple_plots(parsed_results$selected_primer_pairs, "single_primer_report")
plots$amplicon_plot

library(multiplexILP)

# Generate simulated data
primer_pair_data <- generate_primer_pair_data(
  n_pairs = 500,
  n_loci = 6,
  num_channels = 2,
  size_gap = 10,
  seed = 1123
)

View( as.data.frame(primer_pair_data[1:12]))

# Generate ILP model
ilp_result <- gen_ilp_primer_pair(primer_pair_data, lower_bound = 0)

# Solve with SCIP
unlink("sol.txt")
system(paste0("scip -f ", ilp_result$lp_file, " -l sol.txt"))

# Parse results
parsed_results <- parse_scip_locus_C_kt_solution(
  scip_output_file = "sol.txt",
  primers_info_file = ilp_result$primers_info_file,
  output_prefix = "primer_pair_analysis"
)

# Visualize
plots <- generate_simple_plots(parsed_results$selected_primers, "primer_pair_report")
plots$amplicon_plot

library(multiplexILP)
library(stringr)

gen_cons<-function(n=20, percent=0.8){
  
  result <- expand.grid(i = 1:(n-1), j = 1:(n-1));
  result <- result[result$j > result$i, ];
  x <- result$i;
  y <- result$j;
  
  n_keep <- floor(length(x) * percent)  # 保留90%
  keep_idx <- sample(length(x), n_keep)  # 随机选择要保留的位置
  
  x_trimmed <- x[keep_idx]
  y_trimmed <- y[keep_idx]
  
  cons<-str_c("X", x_trimmed," + X",  y_trimmed," <= 1")
  unique(cons)
  
}
# Generate raw constraints randomly
raw_constraints <- gen_cons(n = 60, percent = 0.9)

# Optimize constraints
optimized_cons <- convert_constraints(raw_constraints)

cat("Original:", length(raw_constraints), "constraints\n")
cat("Original (10):", (raw_constraints)[1:10], sep = "\n")


cat("Optimized:", length(optimized_cons), "constraints\n")
cat("Optimized (5):", (optimized_cons)[1:5], sep = "\n")