NGS antibody sequence analysis tool
Absequious uses a Hidden Markov Model to identify antibody variable region domains and calculate high-level diversity statistics. It relies on HMMer for alignments, and is therefore fast enough for analysis of NGS data.
- clone frequencies
- domain annotation (frameworks & CDRs)
- sequence quality annotations (frameshift, stop codons)
python3 -m absequious aln sequences.fa
This will produce 2 files, sequences.fa_reads.csv and sequences.fa_summ.csv. (A different base filename can be specified with the --output_base parameter.) The "reads" file contains every input read, translated and with regions identified, while the "summary" file contains high-level diversity statistics and unique sequences.
- clustering / binning
- liability annotations
- HMMER for domain annotations