On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t
In the last 15 years or so, soft selective sweep mechanisms have been catapulted from a curiosity of little evolutionary importance to a ubiquitous mechanism claimed to explain most adaptive evolution and, in some cases, most evolution. This transformation was aided by a series of articles by Daniel...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Genes |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4425/12/4/527 |
| _version_ | 1797538809357795328 |
|---|---|
| author | Eran Elhaik Dan Graur |
| author_facet | Eran Elhaik Dan Graur |
| author_sort | Eran Elhaik |
| collection | DOAJ |
| description | In the last 15 years or so, soft selective sweep mechanisms have been catapulted from a curiosity of little evolutionary importance to a ubiquitous mechanism claimed to explain most adaptive evolution and, in some cases, most evolution. This transformation was aided by a series of articles by Daniel Schrider and Andrew Kern. Within this series, a paper entitled “Soft sweeps are the dominant mode of adaptation in the human genome” (Schrider and Kern, <i>Mol. Biol. Evolut</i>. <b>2017</b>, <i>34</i>(8), 1863–1877) attracted a great deal of attention, in particular in conjunction with another paper (Kern and Hahn, <i>Mol. Biol. Evolut</i>. <b>2018</b>, <i>35</i>(6), 1366–1371), for purporting to discredit the Neutral Theory of Molecular Evolution (Kimura 1968). Here, we address an alleged novelty in Schrider and Kern’s paper, i.e., the claim that their study involved an artificial intelligence technique called supervised machine learning (SML). SML is predicated upon the existence of a training dataset in which the correspondence between the input and output is known empirically to be true. Curiously, Schrider and Kern did not possess a training dataset of genomic segments known <i>a priori</i> to have evolved either neutrally or through soft or hard selective sweeps. Thus, their claim of using SML is thoroughly and utterly misleading. In the absence of legitimate training datasets, Schrider and Kern used: (1) simulations that employ many manipulatable variables and (2) a system of data cherry-picking rivaling the worst excesses in the literature. These two factors, in addition to the lack of negative controls and the irreproducibility of their results due to incomplete methodological detail, lead us to conclude that all evolutionary inferences derived from so-called SML algorithms (e.g., S/HIC) should be taken with a huge shovel of salt. |
| first_indexed | 2024-03-10T12:36:41Z |
| format | Article |
| id | doaj.art-5db7f03a24624622ba70754f919c482d |
| institution | Directory Open Access Journal |
| issn | 2073-4425 |
| language | English |
| last_indexed | 2024-03-10T12:36:41Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Genes |
| spelling | doaj.art-5db7f03a24624622ba70754f919c482d2023-11-21T14:14:22ZengMDPI AGGenes2073-44252021-04-0112452710.3390/genes12040527On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’tEran Elhaik0Dan Graur1Department of Biology, Lund University, Sölvegatan 35, 22362 Lund, SwedenDepartment of Biology & Biochemistry, University of Houston, Science & Research Building 2, Suite #342, 3455 Cullen Bldv., Houston, TX 77204-5001, USAIn the last 15 years or so, soft selective sweep mechanisms have been catapulted from a curiosity of little evolutionary importance to a ubiquitous mechanism claimed to explain most adaptive evolution and, in some cases, most evolution. This transformation was aided by a series of articles by Daniel Schrider and Andrew Kern. Within this series, a paper entitled “Soft sweeps are the dominant mode of adaptation in the human genome” (Schrider and Kern, <i>Mol. Biol. Evolut</i>. <b>2017</b>, <i>34</i>(8), 1863–1877) attracted a great deal of attention, in particular in conjunction with another paper (Kern and Hahn, <i>Mol. Biol. Evolut</i>. <b>2018</b>, <i>35</i>(6), 1366–1371), for purporting to discredit the Neutral Theory of Molecular Evolution (Kimura 1968). Here, we address an alleged novelty in Schrider and Kern’s paper, i.e., the claim that their study involved an artificial intelligence technique called supervised machine learning (SML). SML is predicated upon the existence of a training dataset in which the correspondence between the input and output is known empirically to be true. Curiously, Schrider and Kern did not possess a training dataset of genomic segments known <i>a priori</i> to have evolved either neutrally or through soft or hard selective sweeps. Thus, their claim of using SML is thoroughly and utterly misleading. In the absence of legitimate training datasets, Schrider and Kern used: (1) simulations that employ many manipulatable variables and (2) a system of data cherry-picking rivaling the worst excesses in the literature. These two factors, in addition to the lack of negative controls and the irreproducibility of their results due to incomplete methodological detail, lead us to conclude that all evolutionary inferences derived from so-called SML algorithms (e.g., S/HIC) should be taken with a huge shovel of salt.https://www.mdpi.com/2073-4425/12/4/527artificial intelligence (AI)supervised machine learning (SML)evolutionary biologymolecular and genome evolutionselective sweepspopulation size |
| spellingShingle | Eran Elhaik Dan Graur On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t Genes artificial intelligence (AI) supervised machine learning (SML) evolutionary biology molecular and genome evolution selective sweeps population size |
| title | On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t |
| title_full | On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t |
| title_fullStr | On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t |
| title_full_unstemmed | On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t |
| title_short | On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t |
| title_sort | on the unfounded enthusiasm for soft selective sweeps iii the supervised machine learning algorithm that isn t |
| topic | artificial intelligence (AI) supervised machine learning (SML) evolutionary biology molecular and genome evolution selective sweeps population size |
| url | https://www.mdpi.com/2073-4425/12/4/527 |
| work_keys_str_mv | AT eranelhaik ontheunfoundedenthusiasmforsoftselectivesweepsiiithesupervisedmachinelearningalgorithmthatisnt AT dangraur ontheunfoundedenthusiasmforsoftselectivesweepsiiithesupervisedmachinelearningalgorithmthatisnt |