The pipeline is developed for classifying the transcriptome with no reference genome.
git clone git@github.com:lizhao007/DenovoAS_Finder.git
BLAST+ and mcl should be installed. Other require python packages should be installed as reminder.
Test.txt and data.fa should be prepared. The test.txt is the alignment result of test.fa and itself by BLAST+, which inculde 12 lines "gene label pident length mismatch gapopen qstart qend sstart send evalue bitscore". "gene" line combines isoform1 and isoform2 with *, and "label" line is consist of 0 or 1, while 0 presents that two isoforms come from different genes and 1 presents the opposite. Other lines produced by "blastn -query test.fa -db test -outfmt 6 -out test.txt -evalue 1e-10". The data.fa is the fasta file contained isoforms which you want to classify.
python DenovoAS_Finder.py test.txt data.fa
Li, Z., Han, L., Luo, Z., and Li, L. Single-molecule long-read sequencing reveals extensive genomic and transcriptomic variation between maize and its wild relative teosinte (Zea mays ssp. parviglumis). Molecular Ecology Resources. https://doi.org/10.1111/1755-0998.13454