camplicon

IN DEVELOPMENT

Find kmers that will act as effective custom amplicon primers

Usage

python camplicon.py <command> <options>

The workflow is split into stages, each of which can be selected with the command argument, or one of two full workflows can be chosen.

kmers

Option	Description
--fg fg_dir, --foreground fg_dir	Directory containing sequences in fasta format
--bg bg_dir, --background bg_dir	Directory containing sequences in fasta format
--kmc kmc_dir, --kmc_dir kmc_dir	Directory containing the KMC executables
--kmer_len kmer_len	Kmer/primer length

primers

Option	Description
--kmer_file kmer_file	Sorted count file produced by KMC
--max_kmers max_kmers	Maximum number of kmers to try (selected at random from candidates). Use 0 to try all kmers.
--freq low_freq	Minimum frequency of kmer to check. Default: most frequent.
--p3 p3_config	Path to Primer3 config directory

filter

Option	Description
--primer_file PRIMER_FILE	Primer file produced by the kmers subcommand
--max_primers max_primers	Maximum number of primers to try (selected at random from candidates). Use 0 to try all primers.
--fg fg_dir, --foreground fg_dir	Directory containing foreground sequences in fasta format
--bg bg_dir, --background bg_dir	Directory containing background sequences in fasta format
--min min_length	Minimum PCR product length
--max max_length	Maximum PCR product length
--ref ref_genome	Genbank file for one of the target sequences to identify context-aware primer locations. File name should be identical except for the file type suffix.
--p3 p3_config	Path to Primer3 config directory

predict

Option	Description
--fg fg_dir, --foreground fg_dir	Directory containing foreground sequences in fasta format
--bg bg_dir, --background bg_dir	Directory containing background sequences in fasta format
--fp FP, --fwd_primer FP	Forward primer sequence
--rp RP, --rev_primer RP	Reverse primer sequence

full

This workflow runs kmers --> primers --> filter --> predict (on the best primer pair)

pfp

This workflow requires a KMC kmer count file and then runs primers --> filter --> predict (on the best primer pair)

How it works

KMC is used to find the unique kmers in a genome, which are then added to a master count of kmers and the number of genomes they appear in uniquely. For now we ignore that some kmers may appear multiple times in a specific genome and therefore be unsuitable as unique primers.

From there, each kmer is checked for viability as a PCR primer with Primer3. All possible pairs are constructed from viable primers and are then checked again by Primer3.

Next, in silico PCR is performed to find the possible products for the in-group of genomes for each primer pair, discarding any that are too long or short.

Finally, BWA is used to align the primers to the out-group genomes, and another round of in silico PCR determines the possible products that might be created.

The output summarises the primer pairs' melting temperatures, number of in-group genomes hit, number of different sequence products produced, information content of products, penalty assigned by Primer3, the mean and stdev in the length of products, the number of out-group genomes hit, the number of different out-group sequence products produced and the overlap between the in-group and out-group products.