Perspective | Published:

Deep learning in biomedicine

Nature Biotechnology volume 36, pages 829838 (2018) | Download Citation

Abstract

Deep learning is beginning to impact biological research and biomedical applications as a result of its ability to integrate vast datasets, learn arbitrarily complex relationships and incorporate existing knowledge. Already, deep learning models can predict, with varying degrees of success, how genetic variation alters cellular processes involved in pathogenesis, which small molecules will modulate the activity of therapeutically relevant proteins, and whether radiographic images are indicative of disease. However, the flexibility of deep learning creates new challenges in guaranteeing the performance of deployed systems and in establishing trust with stakeholders, clinicians and regulators, who require a rationale for decision making. We argue that these challenges will be overcome using the same flexibility that created them; for example, by training deep models so that they can output a rationale for their predictions. Significant research in this direction will be needed to realize the full potential of deep learning in biomedicine.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

References

  1. 1.

    Autonomous vehicles: no drivers required. Nature 518, 20–23 (2015).

  2. 2.

    et al. Human-level control through deep reinforcement learning. Nature 518, 529–533 (2015).

  3. 3.

    et al. Mastering the game of Go without human knowledge. Nature 550, 354–359 (2017).

  4. 4.

    , & Image style transfer using convolutional neural networks. in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016).

  5. 5.

    , & Speech recognition with deep recurrent neural networks. in 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (2013).

  6. 6.

    , & Sequence to sequence learning with neural networks. in. Neural Information Processing Systems 2014, 3104–3112 (2014).

  7. 7.

    , & Deep learning. Nature 521, 436–444 (2015).

  8. 8.

    , , & Deep learning for computational biology. Mol. Syst. Biol. 12, 878 (2016).

  9. 9.

    , , & Machine learning in genomic medicine: a review of computational problems and data sets. Proc. IEEE 104, 176–197 (2016).

  10. 10.

    , , & Applications of deep learning in biomedicine. Mol. Pharm. 13, 1445–1454 (2016).

  11. 11.

    , & Deep learning in bioinformatics. Brief. Bioinform. 18, 851–869 (2017).

  12. 12.

    , & Deep learning in drug discovery. Mol. Inform. 35, 3–14 (2016).

  13. 13.

    et al. An introduction to deep learning on biological sequence data: examples and solutions. Bioinformatics 33, 3685–3690 (2017).

  14. 14.

    et al. Opportunities and obstacles for deep learning in biology and medicine. J. R. Soc. Interface 15, 20170387 (2018).

  15. 15.

    Deep learning in biomedical data science. Annu. Rev. Biomed. Data Sci. 1, 181–205 (2018).

  16. 16.

    An overview of multi-task learning in deep neural networks. Preprint at (2017).

  17. 17.

    , , , & Hierarchical representations for efficient architecture search. Preprint at (2017).

  18. 18.

    , & A survey of transfer learning. J. Big Data 3, 9 (2016).

  19. 19.

    , & ImageNet classification with deep convolutional neural networks. Commun. ACM 60, 84–90 (2017).

  20. 20.

    & Bidirectional recurrent neural networks. IEEE Trans. Signal Process. 45, 2673–2681 (1997).

  21. 21.

    , , & The wake-sleep algorithm for unsupervised neural networks. Science 268, 7761831 (1995).

  22. 22.

    et al. Generative adversarial networks. Preprint at (2014).

  23. 23.

    , , & Unsupervised feature construction and knowledge extraction from genome-wide assays of breast cancer with denoising autoencoders. Pac. Symp. Biocomput. 2015, 132–143 (2015).

  24. 24.

    et al. Automatic chemical design using a data-driven continuous representation of molecules. ACS Cent. Sci. 4, 268–276 (2018).

  25. 25.

    , , & Semi-supervised learning with deep generative models. Preprint at (2014).

  26. 26.

    , , , & Efficient neural architecture search via parameter sharing. Preprint at (2018).

  27. 27.

    A practical Bayesian framework for backpropagation networks. Neural Comput. 4, 448–472 (1992).

  28. 28.

    Bayesian Learning for Neural Networks (Springer, Berlin and Heidelberg, Germany, 1996).

  29. 29.

    & Dropout as a Bayesian approximation: representing model uncertainty in deep learning. Preprint at (2015).

  30. 30.

    Bootstrap methods: another look at the jackknife. Ann. Stat. 7, 1–26 (1979).

  31. 31.

    et al. RNA splicing. The human splicing code reveals new insights into the genetic determinants of disease. Science 347, 1254806 (2015).

  32. 32.

    The mythos of model interpretability. Preprint at (2016).

  33. 33.

    Causal inference in statistics: an overview. Stat. Surv. 3, 96–146 (2009).

  34. 34.

    , , & Not just a black box: learning important features through propagating activation differences. Preprint at (2016).

  35. 35.

    et al. Reports from CAGI: the critical assessment of genome interpretation. Hum. Mutat. 38, 1039–1041 (2017).

  36. 36.

    et al. 10 years of GWAS discovery: biology, function, and translation. Am. J. Hum. Genet. 101, 5–22 (2017).

  37. 37.

    et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet. Med. 17, 405–424 (2015).

  38. 38.

    , , , & Genetic architecture: the shape of the genetic contribution to human traits and disease. Nat. Rev. Genet. 19, 110–124 (2018).

  39. 39.

    , & An expanded view of complex traits: from polygenic to omnigenic. Cell 169, 1177–1186 (2017).

  40. 40.

    , , & Allele-specific transcription factor binding as a benchmark for assessing variant impact predictors. Preprint at bioRxiv (2018).

  41. 41.

    , , & Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nat. Biotechnol. 33, 831–838 (2015).

  42. 42.

    , & Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks. Genome Res. 26, 990–999 (2016).

  43. 43.

    & Predicting effects of noncoding variants with deep learning-based sequence model. Nat. Methods 12, 931–934 (2015).

  44. 44.

    & DanQ: a hybrid convolutional and recurrent deep neural network for quantifying the function of DNA sequences. Nucleic Acids Res. 44, e107 (2016).

  45. 45.

    , , & DeepCpG: accurate prediction of single-cell DNA methylation states using deep learning. Genome Biol. 18, 67 (2017).

  46. 46.

    , , , & TITER: predicting translation initiation sites by deep learning. Bioinformatics 33, i234–i242 (2017).

  47. 47.

    & Massively parallel genetics. Genetics 203, 617–619 (2016).

  48. 48.

    et al. Using deep learning to model the hierarchical structure and function of a cell. Nat. Methods 15, 290–298 (2018).

  49. 49.

    , , , & Exploiting the past and the future in protein secondary structure prediction. Bioinformatics 15, 937–946 (1999).

  50. 50.

    , , & Improving the prediction of protein secondary structure in three and eight classes using recurrent neural networks and profiles. Proteins 47, 228–235 (2002).

  51. 51.

    et al. Convolutional networks on graphs for learning molecular fingerprints. Preprint at (2015).

  52. 52.

    , , , & Molecular graph convolutions: moving beyond fingerprints. J. Comput. Aided Mol. Des. 30, 595–608 (2016).

  53. 53.

    , & Multi-task neural networks for QSAR predictions. Preprint at (2014).

  54. 54.

    , , , & Deep neural nets as a method for quantitative structure-activity relationships. J. Chem. Inf. Model. 55, 263–274 (2015).

  55. 55.

    et al. Massively multitask networks for drug discovery. Preprint at (2015).

  56. 56.

    , & AtomNet: a deep convolutional neural network for bioactivity prediction in structure-based drug discovery. Preprint at (2015).

  57. 57.

    et al. Detecting cancer metastases on gigapixel pathology images. Preprint at (2017).

  58. 58.

    , , , & Deep learning for identifying metastatic breast cancer. Preprint at (2016).

  59. 59.

    & Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology 284, 574–582 (2017).

  60. 60.

    et al. Automated analysis of high-content microscopy data with deep learning. Mol. Syst. Biol. 13, 924 (2017).

  61. 61.

    et al. CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biol. 7, R100 (2006).

  62. 62.

    , & Understanding and confronting our mistakes: the epidemiology of error in radiology and strategies for error reduction. Radiographics 35, 1668–1676 (2015).

  63. 63.

    , & The Sequence Read Archive. Nucleic Acids Res. 39, D19–D21 (2011).

Download references

Acknowledgements

Our perspectives were influenced by conversations with many people, including members of Deep Genomics, B. Andrews, Y. Bengio, B. Blencowe, C. Boone, D. Botstein, C. Francis, A. Heifets, G. Hinton, T. Hughes, P. Hutt, R. Klausner, E. Lander, Y. LeCun, A. Levin, Q. Morris, B. Neale, S. Scherer and J.C. Venter.

Author information

Affiliations

  1. Deep Genomics Inc., MaRS Discovery District, Toronto, Ontario, Canada.

    • Michael Wainberg
    • , Daniele Merico
    • , Andrew Delong
    •  & Brendan J Frey
  2. Department of Computer Science, Stanford University, Stanford, California, USA.

    • Michael Wainberg

Authors

  1. Search for Michael Wainberg in:

  2. Search for Daniele Merico in:

  3. Search for Andrew Delong in:

  4. Search for Brendan J Frey in:

Competing interests

All authors are, or recently were, employees of Deep Genomics, an AI therapeutics company, which is using deep learning to identify the genetic determinants of disease and to develop therapies.

Corresponding author

Correspondence to Brendan J Frey.

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/nbt.4233

Further reading