SACall: a neural network basecaller for Oxford Nanopore sequencing data based on self-attention mechanism.

bash run_caller.sh <model file> <fast5 folder> <signal window length> <output file name>

command parameters
model file: we provide the model.chkpt trained on Klebsiella pneumoniae genome.
fast5 folder: directory of original sequencing files.
signal window length: the length of the signal segment, default: 2048.
output file name: the name of basecalled file.

• python3
• pytorch_gpu v1.0.1
• ctcdecode: https://github.com/parlance/ctcdecode.git