Getting Started

git clone https://github.com/RunpengLuo/PlsChain.git
cd PlsChain && make
# create an index for the plasmid library with k=15
./plschain -i -k 15 -o lib_idx/ backbone.fa promotor.fa peptide.fa gene.fa terminal.fa terminator.fa
# classify the reads against the indexed library
./plschain -q lib_idx/ -o qry_res/ query.fastq.gz
# perform fuzzy match and group the classification
python scripts/plschain_postprocess.py qry_res/ lib_idx/

About PlsChain

PlsChain is an algorithm to classify Oxford Nanopore noisy reads (~5% error rate) sequenced from the plasmid mixtures, it solves the cyclic co-linear chaining problem in the cyclic manner.

Installation

The program is designated for Unix-like system (Linux & MacOS), C compiler, GNU make and zlib development files are required to compile the program.

Run the python script scripts/plschain_postprocess.py for grouping the results with a Python3 environment with no additional library been required.

Program Usage

Usage: plschain -i -k INT -o DIRECTORY FILE1 FILE2 FILE3 ...
       plschain -q DIRECTORY -o DIRECTORY <query.fa>
Options:
    -i            Indexing mode
    -q DIRECTORY  Query mode, index directory
    -k INT        k-mer size [15,32]
    -o DIRECTORY  output directory
    -h            show this message

• FILE1 FILE2 ... consists the library of expression cassettes (with backbone removed), the order should follow the plasmid structure, cyclic order is allowed, e.g., backbone.fa promotor.fa peptide.fa gene.fa terminal.fa terminator.fa.

$python scripts/plschain_postprocess.py
scripts/plschain_postprocess.py <query_dir> <index_dir>

• index_dirrefers to the output directory after running PlsChain with -i indexing mode, and query_dir refers to the output directory after running PlsChain with -q query mode.

Program Output

• <out_dir>/qry_total.csv and <out_dir>/qry_total.fuzzy.csv stores the classification result per read with and without fuzzy match opertaions. Each row consists read name, followed by the ordered list of classified components. * indicates the corresponding component is not decided by PlsChain. fail indicates unclassified record. contamination indicates the filtered unclassified record as contamination based on read length.

• <out_dir>/qry_total.group.csv and <out_dir>/qry_total.group.fuzzy.csv stores the grouped results based on <out_dir>/qry_total.csv and <out_dir>/qry_total.fuzzy.csv, respectively.

Simulation

PlsChain also provides a simulation script plschain_simulator.py that simulates sequencing data from a library of expression cassettes. plschain_simulator.py operates in three modes: sub_sampling, all_sampling, and real_sampling. plschain_simulator.py takes a configuration file and generate index and FASTA file. An example of the configuration file is provided at scripts/sim_conf.txt. Please check out the script and provided example for detailed explanation.