memory efficient de novo detection of het snp kmers using counting bloom filters.
Install requirements: rust ver 1.3 or later. Also clang for the htslib dependency. I am thinking about getting rid of this as it is the entire cause of any build problems here. If you do not have rust
curl https://sh.rustup.rs -sSf | sh
echo 'export PATH=~/.cargo/bin:$PATH' >> ~/.bashrc
source ~/.bashrc
which cargo
For OSX only you will need xz for the htslib dependency. Linux should have the required libraries by default. You can get this with homebrew
brew install xz
then find where it installed, and put the include path on your CFLAGS
export CFLAGS='-I/path/to/xz/<version>/include'
or add that to your .bashrc and source it
Then you should be able to clone and install the project.
git clone git@github.com:wheaton5/het_snp_kmers.git
cd het_snp_kmers
cargo build --release
./target/release/het_snp_kmers -h
het_snp_kmers 1.0
Haynes Heaton <whheaton@gmail.com>
Finds kmer pairs that are different in the middle base and each have roughly haploid coverage. Meant for illumina data
as an initial step for de novo phasing.
USAGE:
het_snp_kmers [OPTIONS] --estimated_kmers <estimated_kmers> --inputs <inputs>... --max_coverage <max_coverage> --min_coverage <min_coverage> --output_full_hist <output_full_hist>
FLAGS:
-h, --help Prints help information
-V, --version Prints version information
OPTIONS:
--estimated_kmers <estimated_kmers>
estimated total unique kmers. good rule of thumb is roughly 2 * genome size
-i, --inputs <inputs>...
input sequence files (fastq,fasta can be gzipped, sam, bam) from which to find het snp kmers
-k, --kmer_size <kmer_size> kmer size to use, defaults to 21
--max_coverage <max_coverage> max coverage for each kmer of the pair
--max_error <max_error>
max count of the other two kmers with middle base changed. For best results best to be strict and use 0.
--max_total_coverage <max_total_coverage> max sum of all kmers with middle base changed
--min_coverage <min_coverage> min coverage for each kmer of the pair
--output_full_hist <output_full_hist> file name for full kmer histogram
-t, --threads <threads> number of threads to use, defaults to 1
Example on made up small test data
./target/release/het_snp_kmers --inputs test/data/test.fastq.gz --estimated_kmers 20 --min_coverage 4 --max_coverage 10 --max_total_coverage 80 --max_error 0 --output_full_hist hist.tsv
counting bloom filter created, detecting het kmers
AAAAAAGGGGACCCCCTTTTT 7 AAAAAAGGGGGCCCCCTTTTT 5