Using information theory to decode network coevolution

doi:10.1126/science.abc6344

GSTDTAP > 气候变化

DOI	10.1126/science.abc6344
	Using information theory to decode network coevolution
	Ricard Solé
	2020-06-19
发表期刊	Science
出版年	2020
英文摘要	Walking through a forest, you spot a colorful butterfly larva crawling and munching on a leaf—nothing unusual, just one scene in a calm ecological play. And yet, a massive “arms race” rages between plants and their herbivores ([ 1 ][1], [ 2 ][2]), spurred by information and misinformation transfer. Chemical signals play the role of communicators in a channel that joins each pair of interacting partners. The mechanisms that drive coevolution of these chemically mediated webs have been an active area of research, but a satisfactory theory has yet to be established. On page 1377 of this issue, Zu et al. ([ 3 ][3]) describe a new application of information theory to coevolutionary dynamics in animal-plant networks. Many scholars have explored the role of information in ecology and evolution ([ 4 ][4]) since the 1950s, in the wake of Claude Shannon's groundbreaking theory of information and communication ([ 5 ][5]). But the actual function of information, particularly in an evolutionary context, often has been obscured by insufficient data and the lack of a proper mapping of the links between information and fitness ([ 6 ][6]). Zu et al. made use of plants' secondary metabolism (which forms metabolites not involved in plant growth or development) to couple two bipartite networks, namely the animal-plant (AP) one (i.e., who eats what) and the plant–volatile organic compounds (PV) one [i.e., what volatile organic compounds (VOCs) plants generate]. The authors gathered data from a tropical dry forest; insect larvae were collected from leaves, and trophic interactions were confirmed in the laboratory. The VOC repertoire was retrieved from leaves in the field with a method that characterizes the chemical profile around each leaf. However, the way the authors built a communication channel between the AP and PV networks departed from Shannon's approach in a substantial way. The classical theory considers a sender and a receiver trying to communicate using a given code sent through a noisy channel (Morse code in the old telegraph system would be one example). For each signal sent, there is a probability that it will be misunderstood by the receiver. Not surprisingly, a great deal of standard information theory deals with finding ways of optimizing communication efficiency. However, in an ecological system in which interactions are highly asymmetric (one species eats the other), the needs at the ends of the (coevolving) communication channel are in clear conflict: Insects need to faithfully identify what leaves are edible, and plants need to avoid being identified as edible. The aim of Zu et al. was to connect information-level descriptions of the two networks through a simple coevolutionary model that successfully solves the problem of defining fitness values for both plants and insects. The rules of the model simulate genetic mutations that influence pairwise interactions within the two bipartite networks. In the simulation, a given PV pair was chosen at random. With a certain probability, the connection between the plant and the chemical signal was added (if there was no link) or removed (if there was one), thus increasing or decreasing the VOC repertoire, respectively. A fitness value, F P, was then computed using one of Shannon's entropic measures, which weights the uncertainty associated with chemical recognition by animals. If the new fitness value, F P′, is larger than the original (before the PV choice), the change is accepted, and the VOC repertoire is expanded. This feature of Zu et al. 's coevolutionary model is particularly notable, as communication networks usually are intended to increase channel efficiency through noise reduction (maximizing information transfer between sender and receiver). In their model, however, changes in a plant's VOC repertoire expanded the chemical diversity of volatiles that favor the confusion of the herbivores or, in Shannon's theory, that increase the herbivores' uncertainty about what plant choices they can make. ![Figure][7] Information between an herbivore (such as a caterpillar) and plant is exchanged through volatile chemical signals, a communication that drives coevolutionary arms races. PHOTO: ANTONIOLÓPEZ-CARRETERO The same mutation process was then applied to the herbivore-plant matrix, but now the chosen pairs were AP (insect-plant) duos. These pairs were also added or removed using a fitness measure. In this case, the selection pressure takes place in the opposite direction, toward reduction of the range of alternative chemical signals. Fitness is thus defined from another entropic measure that weights the improvement of insect specialization obtained from the animal-VOC matrix, which is the product of information transfer between the PV and the AP networks. In this way, two optimization constraints—plant VOC production and insect specialization—must be satisfied simultaneously. With their new model, Zu et al. calculated several information measures from sets of interactions between lists of pairwise exchanges. The authors also extracted probabilities of interactions as well as so-called conditional probabilities (e.g., the probability that a given insect will recognize a specific plant). The sets of probabilities were used to estimate the efficiency of coding and decoding strategies. These measures and the predicted network structure matched very well with those estimated from the field data, including the dense set of connections in the PV data as well as the pattern of insect specialization in the AP set. These results were independent of initial conditions, which included both sparse and dense interaction webs. Such agreement supports the proposal that coevolutionary dynamics is at play and illuminates how VOCs influence the fitness of both plants and herbivores ([ 7 ][8]). Both matter (biomass) and energy (metabolism) are key to our understanding of life. Information stored in DNA and signals transmitted between individuals are major driving forces in biology ([ 8 ][9]). In ecology, information theory has delivered testable predictions in the study of energy-flow networks ([ 9 ][10]) and in characterizing the ecological specialization of plant-pollinator networks ([ 10 ][11]). The theoretical framework presented by Zu et al. explicitly connects coevolutionary dynamics and network architecture. This framework can be applied to microbial VOC–based communication associated with plants ([ 11 ][12]) or the evolutionary dynamics of the rhizosphere, where a diverse range of signaling processes connect plants, fungi, and microorganisms ([ 12 ][13]). In a general context, this formal approach to network coevolution could aid efforts to formulate a theory of ecosystem learning through coevolution ([ 13 ][14]). Furthermore, the opposing pressures that shape the evolution of language networks—which is closely related to Zu et al. 's modeling efforts—also include simultaneous minimization of communication costs and generate statistical distributions that reveal universal laws in language organization ([ 14 ][15]). 1. [↵][16]1. L. M. Schoonhoven, 2. J. J. A. van Loon, 3. M. Dicke , Insect-Plant Biology (Oxford Univ. Press, 2005). 2. [↵][17]1. G. A. Howe, 2. G. Jander , Annu. Rev. Plant Biol. 59, 41 (2008). [OpenUrl][18][CrossRef][19][PubMed][20][Web of Science][21] 3. [↵][22]1. P. Zu et al ., Science 368, 1377 (2020). [OpenUrl][23][Abstract/FREE Full Text][24] 4. [↵][25]1. M. I. O'Connor et al ., Front. Ecol. Evol. 7, 219 (2019). [OpenUrl][26][CrossRef][27] 5. [↵][28]1. C. E. Shannon, 2. W. Weaver , The Mathematical Theory of Communication (Univ. of Illinois Press, 1949). 6. [↵][29]1. M. C. Donaldson-Matasci, 2. C. T. Bergstrom, 3. M. Lachmann , Oikos 119, 219 (2010). [OpenUrl][30][CrossRef][31][PubMed][32][Web of Science][33] 7. [↵][34]1. I. T. Baldwin , Curr. Biol. 20, R392 (2010). [OpenUrl][35][CrossRef][36][PubMed][37] 8. [↵][38]1. M. van Baalen , Interface Focus 3, 20130030 (2013). [OpenUrl][39][CrossRef][40] 9. [↵][41]1. R. E. Ulanowicz , Entropy 21, 949 (2019). [OpenUrl][42] 10. [↵][43]1. N. Blüthgen et al ., Curr. Biol. 17, 341 (2007). [OpenUrl][44][CrossRef][45][PubMed][46][Web of Science][47] 11. [↵][48]1. V. Bitas, 2. H.-S. Kim, 3. J. W. Bennett, 4. S. Kang , Mol. Plant Microbe Interact. 26, 835 (2013). [OpenUrl][49][CrossRef][50][PubMed][51] 12. [↵][52]1. V. Venturi, 2. C. Keel , Trends Plant Sci. 21, 187 (2016). [OpenUrl][53][CrossRef][54] 13. [↵][55]1. D. A. Power et al ., Biol. Direct 10, 69 (2015). [OpenUrl][56][CrossRef][57][PubMed][58] 14. [↵][59]1. B. Corominas-Murtra, 2. J. Fortuny, 3. R. V. Solé , Phys. Rev. E 83, 036115 (2011). [OpenUrl][60] Acknowledgments: R.S. is supported by the Generalitat de Catalunya and the Santa Fe Institute. [1]: #ref-1 [2]: #ref-2 [3]: #ref-3 [4]: #ref-4 [5]: #ref-5 [6]: #ref-6 [7]: pending:yes [8]: #ref-7 [9]: #ref-8 [10]: #ref-9 [11]: #ref-10 [12]: #ref-11 [13]: #ref-12 [14]: #ref-13 [15]: #ref-14 [16]: #xref-ref-1-1 "View reference 1 in text" [17]: #xref-ref-2-1 "View reference 2 in text" [18]: {openurl}?query=rft.jtitle%253DAnnual%2Breview%2Bof%2Bplant%2Bbiology%26rft.stitle%253DAnnu%2BRev%2BPlant%2BBiol%26rft.aulast%253DConn%26rft.auinit1%253DE.%2BE.%26rft.volume%253D59%26rft.spage%253D41%26rft.epage%253D66%26rft.atitle%253DPlant%2Bimmunity%2Bto%2Binsect%2Bherbivores.%26rft_id%253Dinfo%253Adoi%252F10.1146%252Fannurev.arplant.59.032607.092825%26rft_id%253Dinfo%253Apmid%252F18031220%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [19]: /lookup/external-ref?access_num=10.1146/annurev.arplant.59.032607.092825&link_type=DOI [20]: /lookup/external-ref?access_num=18031220&link_type=MED&atom=%2Fsci%2F368%2F6497%2F1315.atom [21]: /lookup/external-ref?access_num=000256593200003&link_type=ISI [22]: #xref-ref-3-1 "View reference 3 in text" [23]: {openurl}?query=rft.jtitle%253DScience%26rft.stitle%253DScience%26rft.aulast%253DZu%26rft.auinit1%253DP.%26rft.volume%253D368%26rft.issue%253D6497%26rft.spage%253D1377%26rft.epage%253D1381%26rft.atitle%253DInformation%2Barms%2Brace%2Bexplains%2Bplant-herbivore%2Bchemical%2Bcommunication%2Bin%2Becological%2Bcommunities%26rft_id%253Dinfo%253Adoi%252F10.1126%252Fscience.aba2965%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [24]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjEzOiIzNjgvNjQ5Ny8xMzc3IjtzOjQ6ImF0b20iO3M6MjM6Ii9zY2kvMzY4LzY0OTcvMTMxNS5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30= [25]: #xref-ref-4-1 "View reference 4 in text" [26]: {openurl}?query=rft.jtitle%253DFront.%2BEcol.%2BEvol.%26rft.volume%253D7%26rft.spage%253D219%26rft_id%253Dinfo%253Adoi%252F10.3389%252Ffevo.2019.00219%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [27]: /lookup/external-ref?access_num=10.3389/fevo.2019.00219&link_type=DOI [28]: #xref-ref-5-1 "View reference 5 in text" [29]: #xref-ref-6-1 "View reference 6 in text" [30]: {openurl}?query=rft.jtitle%253DOikos%26rft.volume%253D119%26rft.spage%253D219%26rft_id%253Dinfo%253Adoi%252F10.1111%252Fj.1600-0706.2009.17781.x%26rft_id%253Dinfo%253Apmid%252F25843980%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [31]: /lookup/external-ref?access_num=10.1111/j.1600-0706.2009.17781.x&link_type=DOI [32]: /lookup/external-ref?access_num=25843980&link_type=MED&atom=%2Fsci%2F368%2F6497%2F1315.atom [33]: /lookup/external-ref?access_num=000275557500004&link_type=ISI [34]: #xref-ref-7-1 "View reference 7 in text" [35]: {openurl}?query=rft.jtitle%253DCurrent%2Bbiology%2B%253A%2B%2BCB%26rft.stitle%253DCurr%2BBiol%26rft.aulast%253DBaldwin%26rft.auinit1%253DI.%2BT.%26rft.volume%253D20%26rft.issue%253D9%26rft.spage%253DR392%26rft.epage%253DR397%26rft.atitle%253DPlant%2Bvolatiles.%26rft_id%253Dinfo%253Adoi%252F10.1016%252Fj.cub.2010.02.052%26rft_id%253Dinfo%253Apmid%252F20462477%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [36]: /lookup/external-ref?access_num=10.1016/j.cub.2010.02.052&link_type=DOI [37]: /lookup/external-ref?access_num=20462477&link_type=MED&atom=%2Fsci%2F368%2F6497%2F1315.atom [38]: #xref-ref-8-1 "View reference 8 in text" [39]: {openurl}?query=rft.jtitle%253DInterface%2BFocus%26rft_id%253Dinfo%253Adoi%252F10.1098%252Frsfs.2013.0030%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [40]: /lookup/external-ref?access_num=10.1098/rsfs.2013.0030&link_type=DOI [41]: #xref-ref-9-1 "View reference 9 in text" [42]: {openurl}?query=rft.jtitle%253DEntropy%26rft.volume%253D21%26rft.spage%253D949%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [43]: #xref-ref-10-1 "View reference 10 in text" [44]: {openurl}?query=rft.jtitle%253DCurrent%2Bbiology%2B%253A%2B%2BCB%26rft.stitle%253DCurr%2BBiol%26rft.aulast%253DBluthgen%26rft.auinit1%253DN.%26rft.volume%253D17%26rft.issue%253D4%26rft.spage%253D341%26rft.epage%253D346%26rft.atitle%253DSpecialization%252C%2Bconstraints%252C%2Band%2Bconflicting%2Binterests%2Bin%2Bmutualistic%2Bnetworks.%26rft_id%253Dinfo%253Adoi%252F10.1016%252Fj.cub.2006.12.039%26rft_id%253Dinfo%253Apmid%252F17275300%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [45]: /lookup/external-ref?access_num=10.1016/j.cub.2006.12.039&link_type=DOI [46]: /lookup/external-ref?access_num=17275300&link_type=MED&atom=%2Fsci%2F368%2F6497%2F1315.atom [47]: /lookup/external-ref?access_num=000244463100025&link_type=ISI [48]: #xref-ref-11-1 "View reference 11 in text" [49]: {openurl}?query=rft.jtitle%253DMol.%2BPlant%2BMicrobe%2BInteract.%26rft.volume%253D26%26rft.spage%253D835%26rft_id%253Dinfo%253Adoi%252F10.1094%252FMPMI-10-12-0249-CR%26rft_id%253Dinfo%253Apmid%252F23581824%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [50]: /lookup/external-ref?access_num=10.1094/MPMI-10-12-0249-CR&link_type=DOI [51]: /lookup/external-ref?access_num=23581824&link_type=MED&atom=%2Fsci%2F368%2F6497%2F1315.atom [52]: #xref-ref-12-1 "View reference 12 in text" [53]: {openurl}?query=rft.jtitle%253DTrends%2BPlant%2BSci.%26rft.volume%253D21%26rft.spage%253D187%26rft_id%253Dinfo%253Adoi%252F10.1016%252Fj.tplants.2016.01.005%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [54]: /lookup/external-ref?access_num=10.1016/j.tplants.2016.01.005&link_type=DOI [55]: #xref-ref-13-1 "View reference 13 in text" [56]: {openurl}?query=rft.jtitle%253DBiol.%2BDirect%26rft.volume%253D10%26rft.spage%253D69%26rft_id%253Dinfo%253Adoi%252F10.1186%252Fs13062-015-0094-1%26rft_id%253Dinfo%253Apmid%252F26643685%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [57]: /lookup/external-ref?access_num=10.1186/s13062-015-0094-1&link_type=DOI [58]: /lookup/external-ref?access_num=26643685&link_type=MED&atom=%2Fsci%2F368%2F6497%2F1315.atom [59]: #xref-ref-14-1 "View reference 14 in text" [60]: {openurl}?query=rft.jtitle%253DPhys.%2BRev.%2BE%26rft.volume%253D83%26rft.spage%253D036115%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx
领域	气候变化 ; 资源环境
URL	查看原文
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.173/C666/handle/2XK7JSWQ/276687
专题	气候变化资源环境科学
推荐引用方式 GB/T 7714	Ricard Solé. Using information theory to decode network coevolution[J]. Science,2020.
APA	Ricard Solé.(2020).Using information theory to decode network coevolution.Science.
MLA	Ricard Solé."Using information theory to decode network coevolution".Science (2020).