FastSK: Fast Sequence Analysis with Gapped String Kernels

• Published in: bioRxiv
• Main Authors: Blakely, Derrick, Collins, Eamon, Singh, Ritambhara, Norton, Andrew Patrick, Lanchantin, Jack, Qi, Yanjun
• Format: Paper
• Language: English
• Published: Cold Spring Harbor Cold Spring Harbor Laboratory Press 30.06.2020; Cold Spring Harbor Laboratory
• Edition: 1.2
• Subjects: Algorithms; Binding sites; Bioinformatics; Datasets; Decomposition; Deoxyribonucleic acid; DNA; Homology; Kernels; Neural networks; Nucleotide sequence; Sequence analysis
• ISSN: 2692-8205; 2692-8205
• DOI: 10.1101/2020.04.21.053975

Gapped k-mer kernels with Support Vector Machines (gkm-SVMs) have achieved strong predictive performance on regulatory DNA sequences on modestly-sized training sets. However, existing gkm-SVM algorithms suffer from the slow kernel computation time, as they depend exponentially on the sub-sequence feature-length, number of mismatch positions, and the task's alphabet size. In this work, we introduce a fast and scalable algorithm for calculating gapped k-mer string kernels. Our method, named FastSK, uses a simplified kernel formulation that decomposes the kernel calculation into a set of independent counting operations over the possible mismatch positions. This simplified decomposition allows us to devise a fast Monte Carlo approximation that rapidly converges. FastSK can scale to much greater feature lengths, allows us to consider more mismatches, and is performant on a variety of sequence analysis tasks. On 10 DNA transcription factor binding site (TFBS) prediction datasets, FastSK consistently matches or outperforms the state-of-the-art gkmSVM-2.0 algorithms in AUC, while achieving average speedups in kernel computation of 100 times and speedups of 800 times for large feature lengths. We further show that FastSK outperforms character-level recurrent and convolutional neural networks across all 10 TFBS tasks. We then extend FastSK to 7 English medical named entity recognition datasets and 10 protein remote homology detection datasets. FastSK consistently matches or outperforms these baselines. Our algorithm is available as a Python package and as C++ source code. (Available for download at https://github.com/Qdata/FastSK/. Install with the command make or pip install) Competing Interest Statement The authors have declared no competing interest. Footnotes * We missed two co-authors in the original submission. * https://github.com/QData/FastSK