Maria C Figueroa, Daniel D Gregory, Kenneth H Williford, David J Fike, Timothy W Lyons
{"title":"利用黄铁矿中的硫同位素和微量元素数据探索早期地球和火星生物特征的机器学习方法。","authors":"Maria C Figueroa, Daniel D Gregory, Kenneth H Williford, David J Fike, Timothy W Lyons","doi":"10.1089/ast.2024.0019","DOIUrl":null,"url":null,"abstract":"<p><p>We propose a novel approach to identify the origin of pyrite grains and distinguish biologically influenced sedimentary pyrite using combined <i>in situ</i> sulfur isotope (δ<sup>34</sup>S) and trace element (TE) analyses. To classify and predict the origin of individual pyrite grains, we applied multiple machine-learning algorithms to coupled δ<sup>34</sup>S and TE data from pyrite grains formed from diverse sedimentary, hydrothermal, and metasomatic processes across geologic time. Our unsupervised classification algorithm, K-means++ cluster analysis, yielded six classes based on the formation environment of the pyrite: sedimentary, low temperature hydrothermal, medium temperature, polymetallic hydrothermal, high temperature, and large euhedral. We tested three supervised models (random forest [RF], Naïve Bayes, k-nearest neighbors), and RF outperformed the others in predicting pyrite formation type, achieving a precision (area under the ROC curve) of 0.979 ± 0.005 and an overall average class accuracy of 0.878 ± 0.005. Moreover, we found that coupling TE and δ<sup>34</sup>S data significantly improved the performance of the RF model compared with using either TE or δ<sup>34</sup>S data alone. Our data provide a novel framework for exploring sedimentary rocks that have undergone multiple hydrothermal, magmatic, and metamorphic alterations. Most significant, however, is the demonstrated potential for distinguishing between biogenic and abiotic pyrite in samples from early Earth. This approach could also be applied to the search for potential biosignatures in samples returned from Mars.</p>","PeriodicalId":8645,"journal":{"name":"Astrobiology","volume":null,"pages":null},"PeriodicalIF":3.5000,"publicationDate":"2024-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Machine-Learning Approach to Biosignature Exploration on Early Earth and Mars Using Sulfur Isotope and Trace Element Data in Pyrite.\",\"authors\":\"Maria C Figueroa, Daniel D Gregory, Kenneth H Williford, David J Fike, Timothy W Lyons\",\"doi\":\"10.1089/ast.2024.0019\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>We propose a novel approach to identify the origin of pyrite grains and distinguish biologically influenced sedimentary pyrite using combined <i>in situ</i> sulfur isotope (δ<sup>34</sup>S) and trace element (TE) analyses. To classify and predict the origin of individual pyrite grains, we applied multiple machine-learning algorithms to coupled δ<sup>34</sup>S and TE data from pyrite grains formed from diverse sedimentary, hydrothermal, and metasomatic processes across geologic time. Our unsupervised classification algorithm, K-means++ cluster analysis, yielded six classes based on the formation environment of the pyrite: sedimentary, low temperature hydrothermal, medium temperature, polymetallic hydrothermal, high temperature, and large euhedral. We tested three supervised models (random forest [RF], Naïve Bayes, k-nearest neighbors), and RF outperformed the others in predicting pyrite formation type, achieving a precision (area under the ROC curve) of 0.979 ± 0.005 and an overall average class accuracy of 0.878 ± 0.005. Moreover, we found that coupling TE and δ<sup>34</sup>S data significantly improved the performance of the RF model compared with using either TE or δ<sup>34</sup>S data alone. Our data provide a novel framework for exploring sedimentary rocks that have undergone multiple hydrothermal, magmatic, and metamorphic alterations. Most significant, however, is the demonstrated potential for distinguishing between biogenic and abiotic pyrite in samples from early Earth. This approach could also be applied to the search for potential biosignatures in samples returned from Mars.</p>\",\"PeriodicalId\":8645,\"journal\":{\"name\":\"Astrobiology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.5000,\"publicationDate\":\"2024-10-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Astrobiology\",\"FirstCategoryId\":\"101\",\"ListUrlMain\":\"https://doi.org/10.1089/ast.2024.0019\",\"RegionNum\":3,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ASTRONOMY & ASTROPHYSICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Astrobiology","FirstCategoryId":"101","ListUrlMain":"https://doi.org/10.1089/ast.2024.0019","RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ASTRONOMY & ASTROPHYSICS","Score":null,"Total":0}
引用次数: 0
摘要
我们提出了一种新方法,利用原位硫同位素(δ34S)和痕量元素(TE)组合分析来确定黄铁矿晶粒的来源并区分受生物影响的沉积黄铁矿。为了对单个黄铁矿晶粒进行分类和预测其来源,我们将多种机器学习算法应用于黄铁矿晶粒的δ34S和TE耦合数据,这些黄铁矿晶粒是在不同的沉积、热液和元气过程中形成的,跨越了地质年代。我们的无监督分类算法--K-means++聚类分析--根据黄铁矿的形成环境得出了六个类别:沉积、低温热液、中温、多金属热液、高温和大斜面。我们测试了三种监督模型(随机森林 [RF]、奈夫贝叶斯、k-近邻),RF 在预测黄铁矿形成类型方面优于其他模型,精确度(ROC 曲线下面积)达到 0.979 ± 0.005,总体平均分类精确度为 0.878 ± 0.005。此外,我们还发现,与单独使用 TE 或 δ34S 数据相比,耦合 TE 和 δ34S 数据可显著提高 RF 模型的性能。我们的数据为探索经历了热液、岩浆和变质等多重变化的沉积岩提供了一个新的框架。然而,最重要的是,我们证明了在早期地球样本中区分生物黄铁矿和非生物黄铁矿的潜力。这种方法也可用于在火星返回的样本中寻找潜在的生物特征。
A Machine-Learning Approach to Biosignature Exploration on Early Earth and Mars Using Sulfur Isotope and Trace Element Data in Pyrite.
We propose a novel approach to identify the origin of pyrite grains and distinguish biologically influenced sedimentary pyrite using combined in situ sulfur isotope (δ34S) and trace element (TE) analyses. To classify and predict the origin of individual pyrite grains, we applied multiple machine-learning algorithms to coupled δ34S and TE data from pyrite grains formed from diverse sedimentary, hydrothermal, and metasomatic processes across geologic time. Our unsupervised classification algorithm, K-means++ cluster analysis, yielded six classes based on the formation environment of the pyrite: sedimentary, low temperature hydrothermal, medium temperature, polymetallic hydrothermal, high temperature, and large euhedral. We tested three supervised models (random forest [RF], Naïve Bayes, k-nearest neighbors), and RF outperformed the others in predicting pyrite formation type, achieving a precision (area under the ROC curve) of 0.979 ± 0.005 and an overall average class accuracy of 0.878 ± 0.005. Moreover, we found that coupling TE and δ34S data significantly improved the performance of the RF model compared with using either TE or δ34S data alone. Our data provide a novel framework for exploring sedimentary rocks that have undergone multiple hydrothermal, magmatic, and metamorphic alterations. Most significant, however, is the demonstrated potential for distinguishing between biogenic and abiotic pyrite in samples from early Earth. This approach could also be applied to the search for potential biosignatures in samples returned from Mars.
期刊介绍:
Astrobiology is the most-cited peer-reviewed journal dedicated to the understanding of life''s origin, evolution, and distribution in the universe, with a focus on new findings and discoveries from interplanetary exploration and laboratory research.
Astrobiology coverage includes: Astrophysics; Astropaleontology; Astroplanets; Bioastronomy; Cosmochemistry; Ecogenomics; Exobiology; Extremophiles; Geomicrobiology; Gravitational biology; Life detection technology; Meteoritics; Planetary geoscience; Planetary protection; Prebiotic chemistry; Space exploration technology; Terraforming