WaveNano: a signal-level nanopore base-caller via simultaneous prediction of nucleotide labels and move labels through bi-directional WaveNets
The Oxford MinION nanopore sequencer is the recently appealing third-generation genome sequencing device that is portable and no larger than a cellphone. Despite the benefits of MinION to sequence ultra-long reads in real-time, the high error rate of the existing base-calling methods, especially indels (insertions and deletions), prevents its use in a variety of applications.
In this paper, we show that such indel errors are largely due to the segmentation process on the input electrical current signal from MinION. All existing methods conduct segmentation and nucleotide label prediction in a sequential manner, in which the errors accumulated in the first step will irreversibly influence the final base-calling. We further show that the indel issue can be significantly reduced via accurate labeling of nucleotide and move labels directly from the raw signal, which can then be efficiently learned by a bi-directionalWaveNet model simultaneously through feature sharing. Our bi-directional WaveNet model with residual blocks and skip connections is able to capture the extremely long dependency in the raw signal. Taking the predicted move as the segmentation guidance, we employ the Viterbi decoding to obtain the final base-calling results from the smoothed nucleotide probability matrix.
Our proposed base-caller, WaveNano, achieves good performance on real MinION sequencing data from Lambda phage.
The signal-level nanopore base-callerWaveNano can obtain higher base-calling accuracy, and generate fewer insertions/deletions in the base-called sequences.
Keywordsnanopore sequencing bi-directional WaveNets base-calling third generation sequencing deep learning
We thank Minh Duc Cao and Lachlan J. M. Coin for providing the nanopore sequencing data for the Lambda phage sample. We thank Haotian Teng for providing helpful discussions. This work was supported by the Kind Abdullah Unviersity of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Awards Nos. FCC/1/1976-04, URF/1/2601-01, URF/1/3007-01, URF/1/3412-01 and URF/1/3450-01.
- 15.Van Den Oord, A., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A., and Kavukcuoglu K. (2016) Wavenet: A generative model for raw audio. ArXiv, 1609.03499Google Scholar
- 17.Chung, J., Gulcehre, C., Cho, K. H. and Bengio, Y. (2014) Empirical evaluation of gated recurrent neural networks on sequence modeling. ArXiv, 1412.3555Google Scholar
- 19.He, K., Zhang, X., Ren, S., and Sun, J. (2016) Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las VegasGoogle Scholar
- 23.Pearson, W. R. and Miller, W. (1992) Dynamic programming algorithms for biological sequence comparison. In Methods in Enzymology. pp. 575–601, ElsevierGoogle Scholar
- 28.van den Oord, A., Kalchbrenner, N., Vinyals, O., Espeholt, L., Graves, A., and Kavukcuoglu, K. (2016) Conditional image generation with pixelcnn decoders. In Advances in Neural Information Processing SystemsGoogle Scholar
- 29.Wang S., Sun S., and Xu J. (2016) AUC-maximized deep convolutional neural fields for protein sequence labeling. In Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2016. Lecture Notes in Computer Science, Frasconi P., Landwehr N., Manco G., Vreeken J. (eds) vol 9852. Springer, ChamGoogle Scholar
- 30.Calders T., and Jaroszewicz S. (2007) Efficient AUC optimization for classification. In Knowledge Discovery in Databases: PKDD 2007. Lecture Notes in Computer Science, Kok J. N., Koronacki J., Lopez de Mantaras R., Matwin S., Mladenic D., Skowron A. (eds), vol 4702. Springer, Berlin, HeidelbergGoogle Scholar