SqueezeCall

SqueezeCall: Nanopore basecalling using a Squeezeformer network. In SqueezeCall, convolution layers are used to downsample raw signals. A Squeezeformer network is employed to capture global context. Finally, a CTC decoder generates the DNA sequence by a beam search algorithm. Experiments on multiple species further demonstrate the potential of the Squeezeformer-based model to improve basecalling accuracy and its superiority over a recurrent neural network (RNN)-based model and Transformer-based models.