Quantum processor-inspired machine learning in the biomedical sciences

Richard Y Li; Sharvari Gujja; Sweta R Bajaj; Omar E Gamel; Nicholas Cilfone; Jeffrey R Gulcher; Daniel A Lidar; Thomas W Chittenden

doi:10.1016/j.patter.2021.100246

Quantum processor-inspired machine learning in the biomedical sciences

Patterns (N Y). 2021 Apr 28;2(6):100246. doi: 10.1016/j.patter.2021.100246. eCollection 2021 Jun 11.

Authors

Richard Y Li^{1

2

3}, Sharvari Gujja^{4

5}, Sweta R Bajaj^{4

5}, Omar E Gamel⁴, Nicholas Cilfone⁴, Jeffrey R Gulcher⁶, Daniel A Lidar^{1

3

7

8}, Thomas W Chittenden^{4

5

9}

Affiliations

¹ Department of Chemistry, University of Southern California, 920 Bloom Walk, Los Angeles, CA 90089, USA.
² Computational Biology and Bioinformatics Program, Department of Biological Sciences, University of Southern California, Los Angeles, CA, USA.
³ Center for Quantum Information Science & Technology, University of Southern California, Los Angeles, Boston, CA, USA.
⁴ Computational Statistics and Bioinformatics Group, Genuity AI Research Institute, Genuity Science, 90 Canal Street, Suite 120, Boston, MA 02114, USA.
⁵ Complex Biological Systems Alliance, Medford, MA, USA.
⁶ Cancer Genetics Group, Genuity Science, Boston, MA, USA.
⁷ Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA.
⁸ Department of Physics and Astronomy, University of Southern California, Los Angeles, CA, USA.
⁹ Division of Genetics and Genomics, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA.

Abstract

Recent advances in high-throughput genomic technologies coupled with exponential increases in computer processing and memory have allowed us to interrogate the complex molecular underpinnings of human disease from a genome-wide perspective. While the deluge of genomic information is expected to increase, a bottleneck in conventional high-performance computing is rapidly approaching. Inspired by recent advances in physical quantum processors, we evaluated several unconventional machine-learning (ML) strategies on actual human tumor data, namely "Ising-type" methods, whose objective function is formulated identical to simulated annealing and quantum annealing. We show the efficacy of multiple Ising-type ML algorithms for classification of multi-omics human cancer data from The Cancer Genome Atlas, comparing these classifiers to a variety of standard ML methods. Our results indicate that Ising-type ML offers superior classification performance with smaller training datasets, thus providing compelling empirical evidence for the potential future application of unconventional computing approaches in the biomedical sciences.

Keywords: The Cancer Genome Atlas; cancer genomics; machine learning.