A tool designed to provide preprocessing for FASTQ files. This tool is developed in C++
for single-end non-compressed fastq :
ffq -f example/bad_test.fastq -L 150 -l 4 -t fastq > filtered_test.fastq
it keeps only sequences with a size between 4 and 150 base pairs.
Simply type
make
it generates a binary ffq
.
Check if the program is working:
./ffq -h
should display
fqq 1
ffq -f fastq_path -o path [-L max_length] [-l min_length] [-p num_threads]
f: input file path.
o: output path to write results.
L: Maximum read's length (in nucleotides) to keep (default 150).
l: Minimum read's length to keep (default 3).
p: enable parallel execution with num_threads threads.
t: input file type. Supported types: fasta, fastq, gzfasta, or gzfastq (default: guess).
h: display this help messsage.