Supervised Machine Learning for Population Genetics: A New Paradigm

Abstract

As population genomic datasets grow in size, researchers are faced with the daunting task of making sense of a flood of information. To keep pace with this explosion of data, computational methodologies for population genetic inference are rapidly being developed to best utilize genomic sequence data. In this review we discuss a new paradigm that has emerged in computational population genomics: that of supervised machine learning (ML). We review the fundamentals of ML, discuss recent applications of supervised ML to population genetics that outperform competing methods, and describe promising future directions in this area. Ultimately, we argue that supervised ML is an important and underutilized tool that has considerable potential for the world of evolutionary genomics.

References 90

1. 1.

   Breiman L. Statistical modeling: the two cultures (with comments and a rejoinder by the author). Stat Sci. 2001;16:199-231.

   Not in our library

2. 2.
   A Genomic Map of the Effects of Linked Selection in Drosophila · 2016

   No notice
3. 3.

   Hinton G, et al. Deep neural networks for acoustic modeling in speech recognition: The shared views of four research groups. IEEE Signal Process Mag. 2012;29:82-97.

   Not in our library

4. 4.

   Sebastiani F. Machine learning in automated text categorization. ACM Comput Surv. 2002;34:1-47.

   Not in our library

5. 5.

   Krizhevsky A, Fereira F, et al. Imagenet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems 25. 2012;1097-1105.

   Not in our library

6. 6.
   Deep learning for computational biology · 2016

   No notice
7. 7.

   Byvatov E, Schneider G. Support vector machine applications in bioinformatics. Appl Bioinform. 2003;2:67.

   Not in our library

8. 8.

   Libbrecht MW, Noble WS. Machine learning applications in genetics and genomics. Nat Rev Genet. 2015;16:321-332.

   Not in our library10.1038/nrg3920

9. 9.

   Lin K, et al. Distinguishing positive selection from neutral evolution: boosting the performance of summary statistics. Genetics. 2011;187:229-244.

   Not in our library10.1534/genetics.110.122614

10. 10.

    Mailund T, et al. Estimating divergence time and ancestral effective population size of Bornean and Sumatran orangutan subspecies using a coalescent hidden Markov model. PLoS Genet. 2011;7:e1001319.

    Not in our library10.1371/journal.pgen.1001319

11. 11.

    Kern AD, Haussler D. A population genetic hidden Markov model for detecting genomic regions under selection. Mol Biol Evol. 2010;27:1673-1685.

    Not in our library10.1093/molbev/msq053

12. 12.

    Boitard S, et al. Detecting selective sweeps: a new approach based on hidden Markov models. Genetics. 2009;181:1567-1578.

    Not in our library10.1534/genetics.108.100032

13. 13.

    Pavlidis P, et al. Searching for footprints of positive selection in whole-genome SNP data from nonequilibrium populations. Genetics. 2010;185:907-922.

    Not in our library10.1534/genetics.110.116459

14. 14.

    Pybus M, et al. Hierarchical boosting: a machine-learning framework to detect and classify hard selective sweeps in human populations. Bioinformatics. 2015;31:3946-3952.

    Not in our library10.1093/bioinformatics/btv493

15. 15.

    Ronen R, et al. Learning natural selection from the site frequency spectrum. Genetics. 2013;195:181-193.

    Not in our library10.1534/genetics.113.152587

16. 16.

    Schrider DR, Kern AD. Inferring selective constraint from population genomic data suggests recent regulatory turnover in the human brain. Genome Biol Evol. 2015;7:3511-3528.

    Not in our library10.1093/gbe/evv228

17. 17.
    S/HIC: Robust Identification of Soft and Hard Sweeps Using Machine Learning · 2016

    No notice
18. 18.
    Deep Learning for Population Genetic Inference · 2016

    No notice
19. 19.

    Amit Y, Geman D. Shape quantization and recognition with randomized trees. Neural Comput. 1997;9:1545-1588.

    Not in our library

20. 20.

    Chen D, et al. Blessing of dimensionality: high-dimensional feature and its efficient compression for face verification. 2013;3025-3032.

    Not in our library

21. 21.

    Kotsiantis SB, et al. Supervised machine learning: a review of classification techniques. Artif Intell Rev. 2006;26:159-190.

    Not in our library

22. 22.

    Ghahramani Z, Bousquet O, et al. Unsupervised learning. Advanced Lectures on Machine Learning. 2004;72-112.

    Not in our library

23. 23.

    Novembre J, et al. Genes mirror geography within Europe. Nature. 2008;456:98.

    Not in our library10.1038/nature07331

24. 24.

    Rabiner LR. A tutorial on hidden Markov models and selected applications in speech recognition. Proc IEEE. 1989;77:257-286.

    Not in our library

25. 25.

    Felsenstein J, Churchill GA. A hidden Markov model approach to variation among sites in rate of evolution. Mol Biol Evol. 1996;13:93-104.

    Not in our library10.1093/oxfordjournals.molbev.a025575

26. 26.

    Siepel A, et al. Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res. 2005;15:1034-1050.

    Not in our library10.1101/gr.3715005

27. 27.

    Dutheil JY, et al. Ancestral population genomics: the coalescent hidden Markov model approach. Genetics. 2009;183:259-274.

    Not in our library10.1534/genetics.109.103010

28. 28.
    Genomic Relationships and Speciation Times of Human, Chimpanzee, and Gorilla Inferred from a Coalescent Hidden Markov Model · 2007

    No notice
29. 29.

    Boitard S, et al. Detecting selective sweeps from pooled next-generation sequencing samples. Mol Biol Evol. 2012;29:2177-2186.

    Not in our library10.1093/molbev/mss090

30. 30.

    Liang P, Jordan MI. An asymptotic analysis of generative, discriminative, and pseudolikelihood estimators. 2008;584-591.

    Not in our library

31. 31.

    Anderson J, et al. The more, the merrier: the blessing of dimensionality for learning large Gaussian mixtures. Proc Mach Learn Res. 2014;35:1135-1164.

    Not in our library

32. 32.

    Bengio Y, LeCun Y. Scaling learning algorithms towards AI. Large Scale Kernel Mach. 2007;34:1-41.

    Not in our library

33. 33.

    Poggio T, et al. Why and when can deep-but not shallow-networks avoid the curse of dimensionality: a review. Int J Autom Comput. 2017;14:503-519.

    Not in our library

34. 34.

    Fay JC, Wu CI. Hitchhiking under positive Darwinian selection. Genetics. 2000;155:1405-1413.

    Not in our library10.1093/genetics/155.3.1405

35. 35.

    Fu YX, Li WH. Statistical tests of neutrality of mutations. Genetics. 1993;133:693-709.

    Not in our library10.1093/genetics/133.3.693

36. 36.

    Fu YX. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics. 1997;147:915-925.

    Not in our library10.1093/genetics/147.2.915

37. 37.

    Kelly JK. A test of neutrality based on interlocus associations. Genetics. 1997;146:1197-1206.

    Not in our library10.1093/genetics/146.3.1197

38. 38.

    Kim Y, Nielsen R. Linkage disequilibrium as a signature of selective sweeps. Genetics. 2004;167:1513-1524.

    Not in our library10.1534/genetics.103.025387

39. 39.

    Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics. 1989;123:585-595.

    Not in our library10.1093/genetics/123.3.585

40. 40.
    A Map of Recent Positive Selection in the Human Genome · 2006

    Corrected
41. 41.

    Bengio Y. Learning deep architectures for AI. Found Trends Mach Learn. 2009;2:1-127.

    Not in our library

42. 42.

    Beaumont MA, et al. Approximate Bayesian computation in population genetics. Genetics. 2002;162:2025-2035.

    Not in our library10.1093/genetics/162.4.2025

43. 43.

    Dunham I, et al. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489:57-74.

    Not in our library10.1038/nature11247

44. 44.

    Altshuler DM, et al. An integrated map of genetic variation from 1,092 human genomes. Nature. 2012;491:56-65.

    Not in our library10.1038/nature11632

45. 45.

    Simonsen KL, et al. Properties of statistical tests of neutrality for DNA polymorphism data. Genetics. 1995;141:413-429.

    Not in our library10.1093/genetics/141.1.413

46. 46.

    Meyer KA, et al. Differential gene expression in the human brain is associated with conserved, but not accelerated, noncoding sequences. Mol Biol Evol. 2017;34:1217-1229.

    Not in our library10.1093/molbev/msx076

47. 47.
    Forces Shaping the Fastest Evolving Regions in the Human Genome · 2006

    No notice
48. 48.

    Maynard Smith J, Haigh J. The hitch-hiking effect of a favourable gene. Genet Res. 1974;23:23-35.

    Not in our library

49. 49.
    A Scan for Positively Selected Genes in the Genomes of Humans and Chimpanzees · 2005

    No notice
50. 50.

    Breiman L. Random forests. Mach Learn. 2001;45:5-32.

    Not in our library

51. 51.

    Geurts P, et al. Extremely randomized trees. Mach Learn. 2006;63:3-42.

    Not in our library

52. 52.

    Hermisson J, Pennings PS. Soft sweeps molecular population genetics of adaptation from standing genetic variation. Genetics. 2005;169:2335-2352.

    Not in our library10.1534/genetics.104.036947

53. 53.

    Orr HA, Betancourt AJ. Haldane’s sieve and adaptation from the standing genetic variation. Genetics. 2001;157:875-884.

    Not in our library10.1093/genetics/157.2.875

54. 54.
    Inferring the Joint Demographic History of Multiple Populations from Multidimensional SNP Frequency Data · 2009

    No notice
55. 55.

    Hey J, Nielsen R. Integration within the Felsenstein equation for improved Markov chain Monte Carlo methods in population genetics. Proc Natl Acad Sci. 2007;104:2785-2790.

    Not in our library10.1073/pnas.0611164104

56. 56.

    Li H, Durbin R. Inference of human population history from individual whole-genome sequences. Nature. 2011;475:493-496.

    Not in our library10.1038/nature10231

57. 57.

    Liu X, Fu YX. Exploring population size changes using SNP frequency spectra. Nat Genet. 2015;47:555-559.

    Not in our library10.1038/ng.3254

58. 58.

    Sheehan S, et al. Estimating variable effective population sizes from multiple genomes: a sequentially Markov conditional sampling distribution approach. Genetics. 2013;194:647-662.

    Not in our library10.1534/genetics.112.149096

59. 59.

    Beaumont MA. Approximate Bayesian computation in evolution and ecology. Annu Rev Ecol Evol Syst. 2010;41:379-406.

    Not in our library

60. 60.

    Aeschbacher S, et al. A novel approach for choosing summary statistics in approximate Bayesian computation. Genetics. 2012;192:1027-1047.

    Not in our library10.1534/genetics.112.143164

61. 61.

    Jiang B, et al. Learning summary statistic for approximate Bayesian computation via deep neural network. arXiv. 2015.

    Not in our library

62. 62.

    Blum MG, François O. Non-linear regression models for approximate Bayesian computation. Stat Comput. 2010;20:63-73.

    Not in our library

63. 63.

    Pudlo P, et al. Reliable ABC model choice via random forests. Bioinformatics. 2016;32:859-866.

    Not in our library10.1093/bioinformatics/btv684

64. 64.

    Schrider D, et al. Supervised machine learning reveals introgressed loci in the genomes of Drosophila simulans and D. sechellia. bioRxiv. 2017.

    Not in our library10.1371/journal.pgen.1007341

65. 65.

    Adrian AB, et al. Predictive models of recombination rate variation across the Drosophila melanogaster genome. Genome Biol Evol. 2016;8:2597-2612.

    Not in our library10.1093/gbe/evw181

66. 66.
    New Software for the Fast Estimation of Population Recombination Rates (FastEPRR) in the Genomic Era · 2016

    No notice
67. 67.

    Lin K, et al. A fast estimate for the population recombination rate based on regression. Genetics. 2013;194:473-484.

    Not in our library10.1534/genetics.113.150201

68. 68.

    McVean GA, et al. The fine-scale structure of recombination rate variation in the human genome. Science. 2004;304:581-584.

    Not in our library10.1126/science.1092500

69. 69.

    Ewing GB, Jensen JD. The consequences of not accounting for background selection in demographic inference. Mol Ecol. 2016;25:135-141.

    Not in our library10.1111/mec.13390

70. 70.

    Schrider DR, et al. Effects of linked selective sweeps on demographic inference and model selection. Genetics. 2016;204:1207-1223.

    Not in our library10.1534/genetics.116.190223

71. 71.

    Sermanet P, et al. Overfeat: Integrated recognition, localization and detection using convolutional networks. arXiv. 2013.

    Not in our library

72. 72.

    Graham B. Fractional max-pooling. arXiv. 2014.

    Not in our library

73. 73.

    Yu C-NJ, Joachims T. Learning structural SVMs with latent variables. 2009;1169-1176.

    Not in our library

74. 74.

    Watterson G. On the number of segregating sites in genetical models without recombination. Theor Popul Biol. 1975;7:256-276.

    Not in our library10.1016/0040-5809(75)90020-9

75. 75.

    Hudson RR. Generating samples under a Wright–Fisher neutral model of genetic variation. Bioinformatics. 2002;18:337-338.

    Not in our library10.1093/bioinformatics/18.2.337

76. 76.

    Pedregosa F, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825-2830.

    Not in our library

77. 77.

    Nei M, Li WH. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc Natl Acad Sci. 1979;76:5269-5273.

    Not in our library10.1073/pnas.76.10.5269

78. 78.
    Recent Selective Sweeps in North American Drosophila melanogaster Show Signatures of Soft Sweeps · 2015

    No notice
79. 79.

    Messer PW, Petrov DA. Population genomics of rapid adaptation by soft selective sweeps. Trends Ecol Evol. 2013;28:659-669.

    Not in our library10.1016/j.tree.2013.08.003

80. 80.

    Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995;20:273-297.

    Not in our library

81. 81.

    Quinlan JR. Induction of decision trees. Mach Learn. 1986;1:81-106.

    Not in our library

82. 82.

    Schapire RE. The strength of weak learnability. Mach Learn. 1990;5:197-227.

    Not in our library

83. 83.

    Bishop CM. Neural Networks for Pattern Recognition. 1995.

    Not in our library

84. 84.
    Deep learning · 2015

    No notice
85. 85.

    Schrider DR, et al. Soft shoulders ahead: spurious signatures of soft and partial selective sweeps result from linked hard sweeps. Genetics. 2015;200:267-284.

    Not in our library10.1534/genetics.115.174912

86. 86.

    Jensen JD, et al. Distinguishing between selective sweeps and demography using DNA polymorphism data. Genetics. 2005;170:1401-1410.

    Not in our library10.1534/genetics.104.038224

87. 87.

    Nielsen R, et al. Genomic scans for selective sweeps using SNP data. Genome Res. 2005;15:1566-1575.

    Not in our library10.1101/gr.4252305

88. 88.
    Soft Sweeps Are the Dominant Mode of Adaptation in the Human Genome · 2017

    No notice
89. 89.

    Blum MG, et al. A comparative review of dimension reduction methods in approximate Bayesian computation. Stat Sci. 2013;28:189-208.

    Not in our library

90. 90.

    Marin J-M, et al. ABC random forests for Bayesian parameter inference. arXiv. 2016.

    Not in our library10.1093/bioinformatics/bty867

Cited by papers we hold 9

• Leakage and the reproducibility crisis in machine-learning-based science · 2023
  No notice
• Artificial intelligence for precision medicine in autoimmune liver disease · 2022
  Not checked
• Efficient ancestry and mutation simulation with msprime 1.0 · 2021
  No notice
• Efficiently Summarizing Relationships in Large Samples: A General Duality Between Statistics of Genealogies and Genomes · 2020
  No notice
• An approximate full-likelihood method for inferring selection and allele frequency trajectories from DNA sequence data · 2019
  No notice
• A generic approach towards afterglow luminescent nanoparticles for ultrasensitive in vivo imaging · 2019
  No notice
• Machine learning and complex biological data · 2019
  No notice
• diploS/HIC: An Updated Approach to Classifying Selective Sweeps · 2018
  No notice
• The Unreasonable Effectiveness of Convolutional Neural Networks in Population Genetic Inference · 2018
  No notice

Reads

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.