DNA Storage Simulator

DNA Storage Simulator analyzes and simulates the error profile of Nanopore DNA. It was completed as part of an undergraduate research project at NUS, supervised by Professor Djordje Jevdjic. See our accepted poster at ISPASS '22 for a short summary, and the extended report for details.

The structure is as follows:

• CodeReconstruction: Forked from CodeReconstruction, with modifications to aid testing.

  • real_data_clustered.txt: Stores real data in the form:

    [original strand][\n]
    *****************************[\n]
    [copy][\n]
    [copy][\n]
    ...
    [copy][\n]
    [\n]
    [\n]
    [original strand][\n]
    *****************************[\n]
    [copy][\n]
    [copy][\n]
    ...
    [copy][\n]
    ...
    

  • synth_data_clustered.txt: Stores synthetic data in the same form as real_data_clustered, can be generated via the noisy.py module, or DNASimulator
  • compare.sh: Bash script that runs reconstruction algorithms on real_data_clustered and synth_data_clustered

• Scripts: Contains utility scripts

  • get_ground_from_clustered.py: Parses files of the same form as real_data_clustered.txt above, and generates a file strands.txt containing the original strands only. Run using python get_ground_from_clustered.py
  • noisy.py: Naive simulator that takes in a strands.txt file, sequencing coverage, and error probabilities as input and generates noisy copies of multiple clusters in the same form as real_data_clustered.txt