Command Line Usage

Basic command line usage involves passing a single sequence.

anarcii DVVMTQTPLSLPVSLGDQASISCRSSQSLVHSNGNTYLNWYLQKAGQSPKLLIYKVSNRFSGVPDRFSGSGSGTDFTLKISRVEAEDLGIYFCSQTTHVPPTFGGGTKLEIKR

However all flags can be accessed.

anarcii --seq_type antibody --ncpu 4 --mode speed --output myseq.txt DVVMTQTPLSLPVSLGDQASISCRSSQSLVHSNGNTYLNWYLQKAGQSPKLLIYKVSNRFSGVPDRFSGSGSGTDFTLKISRVEAEDLGIYFCSQTTHVPPTFGGGTKLEIKR

The path to a fasta file can be provided as well as max_seqs_len argument to expand.

anarcii --seq_type antibody --ncpu 4 --mode speed --output myseq.csv --max_seqs_len 2000000 seqs.fasta

All available flags are listed below:

--seq_type, -t - Sequence type (default: antibody). One of antibody, tcr, shark or unknown.

--batch_size, -b - Batch size for processing (default: 32).

--cpu, -c - Run on CPU (default: False, if available runs on a GPU).

--ncpu, -n - Number of CPU threads to use (default: -1, all available).

--mode, -m - Mode for running the model (default: accuracy).

--max_seqs_len, -l - Threshold for batch processing (default: 102,400, processing sequences over this number will run in batch mode).

--output, -o - Specify the output file (must end in .csv or .msgpack).

--version, -V - Show version.

--verbose, -v - Print details of runtime and batch processing.