HTML Gnuplot Python C++ Shell Makefile
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
docs
dymas
experiments
misc/deprec
reports
.gitignore
Makefile
readme.md
requirements.txt

readme.md

Dynamic Mapping Simulator

Introduction

This is repository contains Dynamic Mapping Simulator and supporting information for the paper K. Břinda, V. Boeva, G. Kucherov: Dynamic read mapping and online consensus calling for better variant detection (arXiv:1605.09070).

Dynamic mapping is mapping to a reference, which is continuously corrected according to alignments computed so far. Dynamic Mapping Simulator is a pipeline to simulate dynamic mapping using existing software to evaluate its benefits in comparison to standard static mapping and iterative referencing. For more details, see the paper.

Simulation algorithm

Scheme of the simulation pipeline:

Reads are taken in the following way:

SM = static mapping, DM = dynamic mapping without remapping, DM-remap = dynamic mapping with remapping, IR = iterative referencing

Structure of this repository

  • docs - supplementary materials (S1 and S2 files)
  • dymas - Dynamic Mapping Simulator (Python package)
  • experiments - all runs of all experiments
  • reports - generated reports

Reports

Replication of results

Prerequisities

Experiments

Additional software for reports

  • GNU Parallel
  • LaTeX
  • Inkscape
  • Gnuplot 5

Recommended way of installation using Anaconda

Environment installation:

	conda create -y --name dymas \
	  -c bioconda \
		python==3.4 \
		snakemake samtools git cmake gnuplot ococo numpy biopython pysam==0.8.3

Environment activation:

source activate dymas

Installation of Python packages (in the activated environment)

pip install -r requirements.txt

Replication steps

  1. Install all required software and activate the corresponding Conda environment.
  source activate dymas
  1. Remove computed data
make clean
  1. Download reference genomes
make -C experiments/exp0*
  1. Run experiments (this step will take several hours)
make -C experiments -j 10
  1. Generate reports
make -C reports -j 10