KoNuSeg

KoNuSeg is a C++ tool for fast, efficient repeat detection using k-mer counting and Bloom filters, backed by btllib.

---

Prerequisites & Requirements

Before building KoNuSeg, ensure your system meets the following software and library requirements:

• Compiler: GCC / G++ version 11 or higher (must support C++17 and OpenMP).
• Dependencies:
  • btllib: Library for bioinformatics data structures (Bloom filters, sequence parsing, etc.). This library is included in this repository (version 1.4.9).

---

Installation

Clone the Repository

git clone git@github.com:BilkentCompGen/KoNuSeg.git
cd KoNuSeg

Compilation & Verification

The project includes a Makefile to simplify compilation and testing. Ensure g++-11 is available at /usr/bin/g++-11 on your system, or update the CC variable in the Makefile to match your local compiler path.

1. Build the Executable

To compile the optimized production binary:

make

2. Sanity check

Before running KoNuSeg on your own dataset, execute the included test suite as a sanity check to verify that the binary was built correctly and all bundled dependencies function properly:

make run_test

3. To remove generated binary files:

make clean

Usage

./bin/konuseg.out -fq <fastq> -fa <fasta> -o <bed> [options]

Required Parameters

Parameter	Description
-fq	Path to input FASTQ file
-fa	Path to input FASTA file
-o	Path to output BED file

Optional Parameters

Option	Default	Description
-g	3000000000	Genome size
-s	10010111001110100101110011101001	Seed value
-k	32	K-mer length
-fpr	0.01	False positive rate
-w	5	Filtering window size
-minseg	500	Minimum segment length for merging
-c_thresh	1	Count threshold for merging
-anum	1500000000	Distinct number of k-mers in assembly/reference
-knum	1500000000	Distinct number of k-mers in reads

If -anum and -knum are not known, the size of the genome can be used.

---

Example Command

./bin/konuseg.out \
    -fa sample_ref.fa \
    -fq sample_reads.fq \
    -o output_segments.bed \
    -g 100000000 \
    -s 11111111111111111111111111111111 \
    -k 32 \
    -fpr 0.01 \
    -w 5 \
    -minseg 64 \
    -c_thresh 1