A method for developing PCR primers specific to a group of genome seqeunces (positives) and specifically exclude a second group of genome seqeunces (negatives).
- Annotated reference sequence in fasta and embl formats
- Group of positive genomes
- Group of negative genomes
start_dir=$PWD
conda create -yn rucs_env python=3 parallel primer3-py numpy samtools bwa blast tabulate emboss
conda activate rucs_env
rucs_env_base=$(dirname $(which conda))/../envs/rucs_env/
mkdir -p $rucs_env_base/opt
cd $rucs_env_base/opt
git clone https://bitbucket.org/genomicepidemiology/rucs.git
git clone https://github.com/bogemad/rucs_analysis.git
chmod +x rucs_analysis/* && mv rucs_analysis/* $rucs_env_base/bin
cd rucs
sed -i "s/, '-parse_seqids'//g' primer_core_tools.py
ln -s $PWD/primer_core_tools.py $rucs_env_base/bin/rucs
cd $start_dir
This pipeline can use a large amount of RAM. If you have large genomes present in your positive or negative genomes then the pipeline may fail. You will need ~200 GB of RAM if any of your genomes are > 500 Mbp. Consider using split_fasta.py to separate large genomes into two or more parts (note this will work for negative genomes but not positive ones).
Configuration file can be opened with nano (or any other text editor)...
nano $(dirname $(which conda))/../envs/rucs_env/opt/rucs/settings.default.cjson
no_good_pp to 10000
threads to whatever cpu resources are
Run RUCS using positive (what you want to detect) and negative (what you don’t want to detect) genomes. Add positive genomes to folder positives, negative genomes to folder negatives.
annotated_reference_genome_fasta=<user_input>
conda activate rucs_env
rucs full \
--reference $annotated_reference_genome_fasta \
--positives positives/* --negatives negatives/* \
[--pick_probe]
conda deactivate rucs_env
Find additional options with rucs -h. These can be very useful.
RUCS will generate at least 10 000 primer/probe sets which need to be filtered. With the method below primer sets are tested for uniqueness and location. Primer sets that amplify more than one target in any positive genome are excluded. Primer sets which are not entirely located within a CDS and single exon are also excluded. CDS details are provided with a embl annotated genome of your reference which can be generated from maker annotated genomes using EMBLmyGFF3. You will need to include the fasta of the reference genome to parallel_primersearch.py.
rucs_output_dir=<user_input>
primer_name_prefix=<user_input>
threads=<user_input>
annotated_reference_genome_embl=<user_input>
annotated_reference_genome_fasta=<user_input>
annotated_reference_genome_name=<user_input> # Same as $annotated_reference_genome_fasta without '.fasta' suffix
conda activate rucs_env
cd $rucs_output_dir
extract_primer_fasta.py \
results_best.tsv \
results_best.fa
extract_primers_for_primersearch.py \
results_best.tsv \
results_best.ps_input \
$primer_name_prefix
parallel_primersearch.py \
results_best.ps_input \
$threads positives/* \
$annotated_reference_genome_fasta
get_primer_sets_single_amp_in_all_positives.py \
*.ps_results.tsv \
single_amp_in_all.txt
get_primers_entirely_within_cds.py \
$annotated_reference_genome_embl \
embl \
$annotated_reference_genome_name.ps_results.tsv \
single_amp_in_all.txt \
single_amp_in_cds.tsv
get_single_amp_primer_table.py \
single_amp_in_cds.tsv \
results_best.tsv \
results_best.single_amp_in_all.tsv \
$primer_name_prefix
conda deactivate rucs_env
Primer sets that pass all filters are found in results_best.single_amp_in_all.tsv with primer binding locations and properties.