FMFI UK Master's degree thesis code and sample data

To compile the project in the src folder you need

• cmake > 3.0
• c++ compiler with C++20 support
• libeigen2-dev
• libfmt-dev
• robin-map-dev
• Boost libraries

Run in the src folder

cmake .
make	

Python dependencies: Requirements:

• numpy
• biopython, Nanosim-H for read generation
• matplotlib, networkx for plotting

Run python read_generator.py -n 500000 -d 0.03 -b art -c 30 to generate Art Illumina reads of a random 500000 bases long sequence and its 3% mutated pair with 30x coverage. See program help for more options

Run python read_generator.py -i sequence_1 sequence_2 ... sequence_n -b nanosimh -c 50 to generate reads using Nanosim-H from provided sequences in FASTA format with 50x coverage

Run jf_occurrences read_file_1 ... read_file_n -k 19 to compute a histogram of 19-mers from read files. Omit the -k argument to have the program determine the k value itself. Every file is treated as reads from a separate haplotype when calculating the histogram.

After plotting of histogram, you are queried with entering the range of occurrences from which to export kmers, as well as a sampling rate (in range 0 to 1).

Run ./categorization read_file_1 read_file_2 ... read_file_n --kmers file_with_kmers to categorize reads using a discriminative kmer set in file_with_kmers. See help for more options.

Run the following commands

python scripts/read_generator.py -i ../data/sequence/ecoli-mg1655.fa ../data/sequences/ecoli-UTI89.fa -b art -c 30
python scripts/read_generator.py -i ../data/sequence/ecoli-mg1655.fa ../data/sequences/ecoli-UTI89.fa -b nanosimh -c 75
./jf_occurrences ecoli_MG1655_art_30x.fq ecoli_UTI89_art_30x.fq -k 19

Select a range of 10 to 25, with sampling rate 1

Run

./categorization ecoli_MG1655_nanosimh_75x.fq ecoli_UTI89_nanosimh_75x.fq --kmers 19-mers_10_25_100%.txt -o components

In the components folder you will find the reads assigned in components