Просмотр кода страницы Robustness (evolution)

{{short description|Persistence of a biological trait under uncertain conditions}}

In [[evolutionary biology]], '''robustness''' of a [[biological system]] (also called biological or genetic robustness<ref>{{cite journal|last1=Kitano|first1=Hiroaki|year=2004|title=Biological robustness|journal=Nature Reviews Genetics|volume=5|issue=11|pages=826–37|doi=10.1038/nrg1471|pmid=15520792|s2cid=7644586}}</ref>) is the persistence of a certain characteristic or trait in a system under perturbations or conditions of uncertainty.<ref>{{cite journal|last1=Stelling|first1=Jörg|last2=Sauer|first2=Uwe|last3=Szallasi|first3=Zoltan|last4=Doyle|first4=Francis J.|last5=Doyle|first5=John|year=2004|title=Robustness of Cellular Functions|journal=Cell|volume=118|issue=6|pages=675–85|doi=10.1016/j.cell.2004.09.008|pmid=15369668|s2cid=14214978|author4-link=Francis J. Doyle III|doi-access=free}}</ref><ref>{{cite journal|last1=Félix|first1=M-A|author2-link=Andreas Wagner|last2=Wagner|first2=A|year=2006|title=Robustness and evolution: Concepts, insights and challenges from a developmental model system|journal=Heredity|volume=100|issue=2|pages=132–40|doi=10.1038/sj.hdy.6800915|pmid=17167519|doi-access=free|url=http://hal.archives-ouvertes.fr/docs/00/13/42/96/PDF/FelixWagner_Heredity_2006.pdf}}</ref>  Robustness in development is known as [[canalisation (genetics)|canalization]].<ref>{{cite journal|last1=Waddington|first1=C. H.|year=1942|title=Canalization of Development and the Inheritance of Acquired Characters|journal=Nature|volume=150|issue=3811|pages=563–5|bibcode=1942Natur.150..563W|doi=10.1038/150563a0|s2cid=4127926}}</ref><ref>{{cite journal|last1=De Visser|first1=JA|last2=Hermisson|first2=J|last3=Wagner|first3=GP|last4=Ancel Meyers|first4=L|last5=Bagheri-Chaichian|first5=H|last6=Blanchard|first6=JL|last7=Chao|first7=L|last8=Cheverud|first8=JM|last9=Elena|first9=SF|last10=Fontana|first10=W|last11=Gibson|first11=G|display-authors=8|year=2003|title=Perspective: Evolution and detection of genetic robustness|journal=Evolution; International Journal of Organic Evolution|volume=57|issue=9|pages=1959–72|jstor=3448871|pmid=14575319|last15=Ofria|first15=C|last16=Rice|first16=S. H.|first14=R. C.|first17=G|author18-link=Andreas Wagner|last18=Wagner|first18=A|last19=Whitlock|first19=M. C.|last14=Lewontin|first13=D|last13=Krakauer|first12=T. F.|last12=Hansen|last17=von Dassow|doi=10.1111/j.0014-3820.2003.tb00377.x|s2cid=221736785|doi-access=free}}</ref>  According to the kind of perturbation involved, robustness can be classified as [[mutation]]al, [[environment (biophysical)|environmental]], [[Genetic recombination|recombinational]], or [[behavior]]al robustness ''etc''.<ref name="jaf2011a">{{cite journal|last1=Fernandez-Leon|first1=Jose A.|year=2011|title=Evolving cognitive-behavioural dependencies in situated agents for behavioural robustness|journal=Biosystems|volume=106|issue=2–3|pages=94–110|doi=10.1016/j.biosystems.2011.07.003|pmid=21840371}}</ref><ref name="jaf2011b">{{cite journal|last1=Fernandez-Leon|first1=Jose A.|year=2011|title=Behavioural robustness: A link between distributed mechanisms and coupled transient dynamics|journal=Biosystems|volume=105|issue=1|pages=49–61|doi=10.1016/j.biosystems.2011.03.006|pmid=21466836}}</ref><ref name="jaf2010">{{cite journal|last1=Fernandez-Leon|first1=Jose A.|year=2011|title=Evolving experience-dependent robust behaviour in embodied agents|journal=Biosystems|volume=103|issue=1|pages=45–56|doi=10.1016/j.biosystems.2010.09.010|pmid=20932875}}</ref> Robustness is achieved through the combination of many [[genetics|genetic]] and [[molecular biology|molecular mechanisms]] and can [[Evolution|evolve]] by either direct or indirect [[natural selection|selection]]. Several [[model organism|model systems]] have been developed to experimentally study robustness and its evolutionary consequences.[[File:Robustness cube annotated.svg|thumb|300px|A network of [[genotype]]s linked by mutations. Each genotype is made up of 3 [[gene]]s: a, b & c. Each gene can be one of two [[allele]]s. Lines link different phenotypes by [[mutation]]. The [[phenotype]] is indicated by colour. Genotypes abc, Abc, aBc and abC lie on a [[neutral network (evolution)|neutral network]] since all have the same, dark phenotype. Genotype abc is robust since any single mutation retains the same phenotype. Other genotypes are less robust as mutations change the phenotype (e.g. ABc).]]

==Classification==

===Mutational robustness===
Mutational robustness (also called mutation tolerance) describes the extent to which an organism's phenotype remains constant in spite of [[mutation]].<ref name=Wagner2005 /> Robustness can be empirically measured for several [[genome]]s<ref name="Sanjuán 1975–82">{{cite journal|last=Sanjuán|first=R|title=Mutational fitness effects in RNA and single-stranded DNA viruses: common patterns revealed by site-directed mutagenesis studies.|journal=Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences|date=Jun 27, 2010|volume=365|issue=1548|pages=1975–82|pmid=20478892|doi=10.1098/rstb.2010.0063|pmc=2880115}}</ref><ref>{{cite journal|last=Eyre-Walker|first=A|author2=Keightley, PD |title=The distribution of fitness effects of new mutations.|journal=Nature Reviews Genetics|date=Aug 2007|volume=8|issue=8|pages=610–8|pmid=17637733|doi=10.1038/nrg2146|s2cid=10868777}}</ref> and individual [[gene]]s<ref name="Hietpas 7896–901">{{cite journal|last=Hietpas|first=RT|author2=Jensen, JD |author3=Bolon, DN |title=Experimental illumination of a fitness landscape.|journal=Proceedings of the National Academy of Sciences of the United States of America|date=May 10, 2011|volume=108|issue=19|pages=7896–901|pmid=21464309|doi=10.1073/pnas.1016024108|pmc=3093508|bibcode=2011PNAS..108.7896H|doi-access=free}}</ref> by inducing mutations and measuring what proportion of mutants retain the same [[phenotype]], function or [[fitness (biology)|fitness]]. More generally, robustness corresponds to the neutral band in the [[distribution of fitness effects]] of mutation (i.e. the frequencies of different fitnesses of mutants). Proteins so far investigated have shown a tolerance to mutations of roughly 66% (i.e. two thirds of mutations are neutral).<ref name="Guo 9205–10">{{cite journal|last=Guo|first=HH|author2=Choe, J |author3=Loeb, LA |title=Protein tolerance to random amino acid change.|journal=Proceedings of the National Academy of Sciences of the United States of America|date=Jun 22, 2004|volume=101|issue=25|pages=9205–10|pmid=15197260|doi=10.1073/pnas.0403255101|pmc=438954|bibcode=2004PNAS..101.9205G|doi-access=free}}</ref>

Conversely, measured mutational robustnesses of organisms vary widely. For example, >95% of point mutations in ''[[Caenorhabditis elegans|C. elegans]]'' have no detectable effect<ref name="Davies 1748–1751">{{cite journal|last1=Davies|first1=E. K.|title=High Frequency of Cryptic Deleterious Mutations in Caenorhabditis elegans|journal=Science|date=10 September 1999|volume=285|issue=5434|pages=1748–1751|doi=10.1126/science.285.5434.1748|pmid=10481013|last2=Peters|first2=A. D.|last3=Keightley|first3=P. D.}}</ref> and even 90% of single gene knockouts in ''[[E. coli]]'' are non-lethal.<ref name="Baba 2006 2006.0008">{{cite journal|last=Baba|first=T|author2=Ara, T |author3=Hasegawa, M |author4=Takai, Y |author5=Okumura, Y |author6=Baba, M |author7=Datsenko, KA |author8=Tomita, M |author9=Wanner, BL |author10= Mori, H |title=Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection.|journal=Molecular Systems Biology|year=2006|volume=2|pages=2006.0008|pmid=16738554|doi=10.1038/msb4100050|pmc=1681482|issue=1}}</ref> Viruses, however, only tolerate 20-40% of mutations and hence are much more sensitive to mutation.<ref name="Sanjuán 1975–82"/>

=== Robustness to stochasticity ===
Biological processes at the molecular scale are inherently [[stochastic]].<ref>{{Cite book|title=Stochastic processes in cell biology|last=[[Paul Bressloff|Bressloff, Paul C.]]|isbn=978-3-319-08488-6|location=Cham|oclc=889941610|date = 2014-08-22}}</ref> They emerge from a combination of stochastic events that happen given the physico-chemical properties of molecules. For instance, gene expression is intrinsically noisy. This means that two cells in exactly identical [[Cell type|regulatory states]] will exhibit different [[Messenger RNA|mRNA]] contents.<ref>{{Cite journal|last=Elowitz|first=M. B.|date=2002-08-16|title=Stochastic Gene Expression in a Single Cell|journal=Science|volume=297|issue=5584|pages=1183–1186|doi=10.1126/science.1070919|pmid=12183631|bibcode=2002Sci...297.1183E|s2cid=10845628|url=https://authors.library.caltech.edu/records/wsymf-b6c81/files/ElowitzSOM.pdf?download=1}}</ref><ref>{{cite journal |last1=Blake |first1=William J. |last2=KÆrn |first2=Mads |last3=Cantor |first3=Charles R. |last4=Collins |first4=J. J. |title=Noise in eukaryotic gene expression |journal=Nature |date=April 2003 |volume=422 |issue=6932 |pages=633–637 |doi=10.1038/nature01546 |pmid=12687005 |bibcode=2003Natur.422..633B |s2cid=4347106 }}</ref> The cell population level log-normal distribution of mRNA content<ref>{{cite journal |last1=Bengtsson |first1=M. |last2=Ståhlberg |first2=A |last3=Rorsman |first3=P |last4=Kubista |first4=M |title=Gene expression profiling in single cells from the pancreatic islets of Langerhans reveals lognormal distribution of mRNA levels |journal=Genome Research |date=16 September 2005 |volume=15 |issue=10 |pages=1388–1392 |doi=10.1101/gr.3820805 |pmid=16204192 |pmc=1240081 }}</ref> follows directly from the application of the [[Central limit theorem|Central Limit Theorem]] to the multi-step nature of [[Regulation of gene expression|gene expression regulation]].<ref>{{cite journal |last1=Beal |first1=Jacob |title=Biochemical complexity drives log-normal variation in genetic expression |journal=Engineering Biology |date=1 June 2017 |volume=1 |issue=1 |pages=55–60 |doi=10.1049/enb.2017.0004 |s2cid=31138796 |doi-access=free }}</ref>

===Environmental robustness===
In varying [[environment (biophysical)|environments]], perfect [[adaptation]] to one condition may come at the expense of adaptation to another. Consequently, the total selection pressure on an organism is the average selection across all environments weighted by the percentage time spent in that environment. Variable environment can therefore select for environmental robustness where organisms can function across a wide range of conditions with little change in [[phenotype]] or [[fitness (biology)]]. Some organisms show adaptations to tolerate large changes in temperature, water availability, salinity or food availability. Plants, in particular, are unable to move when the environment changes and so show a range of mechanisms for achieving environmental robustness. Similarly, this can be seen in proteins as tolerance to a wide range of [[solvent]]s, [[ion|ion concentrations]] or [[temperature]]s.

==Genetic, molecular and cellular causes==
[[File:Metabolism diagram.svg|thumb|Core eukaryotic [[metabolism|metabolic network]]. Circles indicate [[metabolite]]s and lines indicate conversions by [[enzyme]]s. Many metabolites can be produced via more than one route, therefore the organism is robust to the loss of some metabolic enzymes]]
Genomes mutate by environmental damage and imperfect replication, yet they display remarkable tolerance. This comes from robustness both at many different levels.

===Organism mutational robustness===
There are many mechanisms that provide genome robustness. For example, [[genetic redundancy]] reduces the effect of mutations in any one copy of a multi-copy gene.<ref>{{cite journal|last=Gu|first=Z|author2=Steinmetz, LM |author3=Gu, X |author4=Scharfe, C |author5=Davis, RW |author6= Li, WH |title=Role of duplicate genes in genetic robustness against null mutations.|journal=Nature|date=Jan 2, 2003|volume=421|issue=6918|pages=63–6|pmid=12511954|doi=10.1038/nature01198|bibcode=2003Natur.421...63G|s2cid=4348693}}</ref> Additionally the [[flux (metabolism)|flux]] through a [[metabolic pathway]] is typically limited by only a few of the steps, meaning that changes in function of many of the enzymes have little effect on fitness.<ref>{{cite journal|last=Kauffman|first=Kenneth J|author2=Prakash, Purusharth |author3=Edwards, Jeremy S |title=Advances in flux balance analysis|journal=Current Opinion in Biotechnology|date=October 2003|volume=14|issue=5|pages=491–496|doi=10.1016/j.copbio.2003.08.001|pmid=14580578}}</ref><ref>{{cite journal|last=Nam|first=H|author2=Lewis, NE |author3=Lerman, JA |author4=Lee, DH |author5=Chang, RL |author6=Kim, D |author7= Palsson, BO |title=Network context and selection in the evolution to enzyme specificity.|journal=Science|date=Aug 31, 2012|volume=337|issue=6098|pages=1101–4|pmid=22936779|doi=10.1126/science.1216861 |pmc=3536066|bibcode=2012Sci...337.1101N}}</ref> Similarly [[metabolic network]]s have multiple alternate pathways to produce many key [[metabolite]]s.<ref>{{cite journal|last=Krakauer|first=DC|author2=Plotkin, JB |title=Redundancy, antiredundancy, and the robustness of genomes.|journal=Proceedings of the National Academy of Sciences of the United States of America|date=Feb 5, 2002|volume=99|issue=3|pages=1405–9|pmid=11818563|doi=10.1073/pnas.032668599|pmc=122203|bibcode=2002PNAS...99.1405K|doi-access=free}}</ref>

===Protein mutational robustness===
Protein mutation tolerance is the product of two main features: the structure of the [[genetic code]] and [[protein structure|protein structural]] robustness.<ref>{{cite journal|last=Taverna|first=DM|author2=Goldstein, RA |title=Why are proteins so robust to site mutations?|journal=Journal of Molecular Biology|date=Jan 18, 2002|volume=315|issue=3|pages=479–84|pmid=11786027|doi=10.1006/jmbi.2001.5226}}</ref><ref>{{cite journal|last=Tokuriki|first=N|author2=Tawfik, DS |title=Stability effects of mutations and protein evolvability.|journal=Current Opinion in Structural Biology|date=Oct 2009|volume=19|issue=5|pages=596–604|pmid=19765975|doi=10.1016/j.sbi.2009.08.003}}</ref> Proteins are resistant to mutations because many sequences can fold into highly similar [[protein fold|structural folds]].<ref>{{cite journal|last=Meyerguz|first=L|author2=Kleinberg, J |author3=Elber, R |title=The network of sequence flow between protein structures.|journal=Proceedings of the National Academy of Sciences of the United States of America|date=Jul 10, 2007|volume=104|issue=28|pages=11627–32|pmid=17596339|doi=10.1073/pnas.0701393104|pmc=1913895|bibcode=2007PNAS..10411627M|doi-access=free}}</ref> A protein adopts a limited ensemble of native conformations because those conformers have lower energy than unfolded and mis-folded states (ΔΔG of folding).<ref>{{cite journal|last=Karplus|first=M|title=Behind the folding funnel diagram.|journal=Nature Chemical Biology|date=Jun 17, 2011|volume=7|issue=7|pages=401–4|pmid=21685880|doi=10.1038/nchembio.565}}</ref><ref>{{cite journal|last=Tokuriki|first=N|author2=Stricher, F |author3=Schymkowitz, J |author4=Serrano, L |author5= Tawfik, DS |title=The stability effects of protein mutations appear to be universally distributed.|journal=Journal of Molecular Biology|date=Jun 22, 2007|volume=369|issue=5|pages=1318–32|pmid=17482644|doi=10.1016/j.jmb.2007.03.069|s2cid=24638570}}</ref> This is achieved by a distributed, internal network of cooperative interactions ([[hydrophobic]], [[polarity (chemistry)|polar]] and [[covalent bond|covalent]]).<ref>{{cite journal|last=Shakhnovich|first=BE|author2=Deeds, E |author3=Delisi, C |author4= Shakhnovich, E |title=Protein structure and evolutionary history determine sequence space topology.|journal=Genome Research|date=Mar 2005|volume=15|issue=3|pages=385–92|pmid=15741509|doi=10.1101/gr.3133605 |pmc=551565|arxiv=q-bio/0404040}}</ref> Protein structural robustness results from few single mutations being sufficiently disruptive to compromise function. Proteins have also evolved to avoid [[protein aggregation|aggregation]]<ref>{{cite journal|last=Monsellier|first=E|author2=Chiti, F |title=Prevention of amyloid-like aggregation as a driving force of protein evolution.|journal=EMBO Reports|date=Aug 2007|volume=8|issue=8|pages=737–42|pmid=17668004|doi=10.1038/sj.embor.7401034|pmc=1978086}}</ref> as partially folded proteins can combine to form large, repeating, insoluble [[Amyloid#CITEREF2002|protein fibrils]] and masses.<ref>{{cite journal|last=Fink|first=AL|title=Protein aggregation: folding aggregates, inclusion bodies and amyloid.|journal=Folding & Design|year=1998|volume=3|issue=1|pages=R9–23|pmid=9502314|doi=10.1016/s1359-0278(98)00002-9|doi-access=free}}</ref> There is evidence that proteins show negative design features to reduce the exposure of aggregation-prone [[beta-sheet]] motifs in their structures.<ref>{{cite journal|last=Richardson|first=JS|author2=Richardson, DC |title=Natural beta-sheet proteins use negative design to avoid edge-to-edge aggregation.|journal=Proceedings of the National Academy of Sciences of the United States of America|date=Mar 5, 2002|volume=99|issue=5|pages=2754–9|pmid=11880627|doi=10.1073/pnas.052706099|pmc=122420|bibcode=2002PNAS...99.2754R|doi-access=free}}</ref>
Additionally, there is some evidence that the [[genetic code]] itself may be optimised such that most point mutations lead to similar amino acids ([[neutral mutation|conservative]]).<ref>{{cite journal|vauthors=Müller MM, Allison JR, Hongdilokkul N, Gaillon L, Kast P, van Gunsteren WF, Marlière P, Hilvert D |title=Directed evolution of a model primordial enzyme provides insights into the development of the genetic code.|journal=PLOS Genetics|year=2013|volume=9|issue=1|article-number=e1003187|pmid=23300488|doi=10.1371/journal.pgen.1003187|pmc=3536711 |doi-access=free }} {{open access}}</ref><ref>{{cite journal|last=Firnberg|first=E|author2=Ostermeier, M |title=The genetic code constrains yet facilitates Darwinian evolution.|journal=Nucleic Acids Research|date=Aug 2013|volume=41|issue=15|pages=7420–8|pmid=23754851|doi=10.1093/nar/gkt536|pmc=3753648}}</ref> Together these factors create a [[distribution of fitness effects]] of mutations that contains a high proportion of neutral and nearly-neutral mutations.<ref name="Hietpas 7896–901"/>

=== Gene expression robustness ===
During [[embryonic development]], gene expression must be tightly controlled in time and space in order to give rise to fully functional organs. Developing organisms must therefore deal with the random perturbations resulting from gene expression stochasticity.<ref name=":0">{{Cite journal|last1=Lagha|first1=Mounia|last2=Bothma|first2=Jacques P.|last3=Levine|first3=Michael|date=2012|title=Mechanisms of transcriptional precision in animal development|journal=Trends in Genetics|language=en|volume=28|issue=8|pages=409–416|doi=10.1016/j.tig.2012.03.006|pmc=4257495|pmid=22513408}}</ref> In [[bilateria]]ns, robustness of gene expression can be achieved via [[Enhancer (genetics)|enhancer]] redundancy. This happens when the expression of a gene under the control of several enhancers encoding the same regulatory logic (ie. displaying binding sites for the same set of [[transcription factor]]s). In ''[[Drosophila melanogaster]]'' such redundant enhancers are often called [[shadow enhancers]].<ref>{{Cite journal|last1=Perry|first1=Michael W.|last2=Boettiger|first2=Alistair N.|last3=Bothma|first3=Jacques P.|last4=Levine|first4=Michael|date=2010|title=Shadow Enhancers Foster Robustness of Drosophila Gastrulation|journal=Current Biology|language=en|volume=20|issue=17|pages=1562–1567|doi=10.1016/j.cub.2010.07.043|pmc=4257487|pmid=20797865}}</ref>

Furthermore, in developmental contexts were timing of gene expression in important for the phenotypic outcome, diverse mechanisms exist to ensure proper gene expression in a timely manner.<ref name=":0" /> [[Poised promoters]] are transcriptionally inactive [[Promoter (genetics)|promoters]] that display [[RNA polymerase II]] binding, ready for rapid induction.<ref>{{cite journal |last1=Zeitlinger |first1=Julia |last2=Stark |first2=Alexander |last3=Kellis |first3=Manolis |last4=Hong |first4=Joung-Woo |last5=Nechaev |first5=Sergei |last6=Adelman |first6=Karen |last7=Levine |first7=Michael |last8=Young |first8=Richard A |title=RNA polymerase stalling at developmental control genes in the Drosophila melanogaster embryo |journal=Nature Genetics |date=11 November 2007 |volume=39 |issue=12 |pages=1512–1516 |doi=10.1038/ng.2007.26 |pmid=17994019 |pmc=2824921 }}</ref> In addition, because not all transcription factors can bind their target site in compacted [[heterochromatin]], [[Pioneer factor|pioneer transcription factors]] (such as ''Zld'' or ''FoxA'') are required to open chromatin and allow the binding of other transcription factors that can rapidly induce gene expression. Open inactive enhancers are call [[poised enhancers]].<ref>{{cite journal |last1=Nien |first1=Chung-Yi |last2=Liang |first2=Hsiao-Lan |last3=Butcher |first3=Stephen |last4=Sun |first4=Yujia |last5=Fu |first5=Shengbo |last6=Gocha |first6=Tenzin |last7=Kirov |first7=Nikolai |last8=Manak |first8=J. Robert |last9=Rushlow |first9=Christine |last10=Barsh |first10=Gregory S. |title=Temporal Coordination of Gene Networks by Zelda in the Early Drosophila Embryo |journal=PLOS Genetics |date=20 October 2011 |volume=7 |issue=10 |article-number=e1002339 |doi=10.1371/journal.pgen.1002339 |pmid=22028675 |pmc=3197689 |doi-access=free }}</ref>

[[Cell competition]] is a phenomenon first described in [[Drosophila melanogaster|Drosophila]]<ref>{{Cite journal|last1=Morata|first1=Ginés|last2=Ripoll|first2=Pedro|date=1975|title=Minutes: Mutants of Drosophila autonomously affecting cell division rate|journal=Developmental Biology|language=en|volume=42|issue=2|pages=211–221|doi=10.1016/0012-1606(75)90330-9|pmid=1116643}}</ref> where mosaic ''Minute'' mutant cells (affecting [[ribosomal protein]]s) in a wild-type background would be eliminated. This phenomenon also happens in the early mouse embryo where cells expressing high levels of ''[[Myc]]'' actively kill their neighbors displaying low levels of ''[[Myc]]'' expression. This results in homogeneously high levels of ''[[Myc]]''.<ref>{{cite journal |last1=Clavería |first1=Cristina |last2=Giovinazzo |first2=Giovanna |last3=Sierra |first3=Rocío |last4=Torres |first4=Miguel |title=Myc-driven endogenous cell competition in the early mammalian embryo |journal=Nature |date=10 July 2013 |volume=500 |issue=7460 |pages=39–44 |doi=10.1038/nature12389 |pmid=23842495 |bibcode=2013Natur.500...39C |s2cid=4414411 }}</ref><ref>{{Cite journal|last1=Sancho|first1=Margarida|last2=Di-Gregorio|first2=Aida|last3=George|first3=Nancy|last4=Pozzi|first4=Sara|last5=Sánchez|first5=Juan Miguel|last6=Pernaute|first6=Barbara|last7=Rodríguez|first7=Tristan A.|date=2013|title=Competitive Interactions Eliminate Unfit Embryonic Stem Cells at the Onset of Differentiation|journal=Developmental Cell|language=en|volume=26|issue=1|pages=19–30|doi=10.1016/j.devcel.2013.06.012|pmc=3714589|pmid=23867226}}</ref>

=== Developmental patterning robustness ===
Patterning mechanisms such as those described by the [[French flag model]] can be perturbed at many levels (production and stochasticity of the diffusion of the morphogen, production of the receptor, stochastic of the [[Signal transduction|signaling cascade]], etc). Patterning is therefore inherently noisy. Robustness against this noise and genetic perturbation is therefore necessary to ensure proper that cells measure accurately positional information. Studies of the [[zebrafish]] [[neural tube]] and antero-posterior patternings has shown that noisy signaling leads to imperfect cell differentiation that is later corrected by transdifferentiation, migration or cell death of the misplaced cells.<ref>{{Cite journal|last1=Xiong|first1=Fengzhu|last2=Tentner|first2=Andrea R.|last3=Huang|first3=Peng|last4=Gelas|first4=Arnaud|last5=Mosaliganti|first5=Kishore R.|last6=Souhait|first6=Lydie|last7=Rannou|first7=Nicolas|last8=Swinburne|first8=Ian A.|last9=Obholzer|first9=Nikolaus D.|last10=Cowgill|first10=Paul D.|last11=Schier|first11=Alexander F.|date=2013|title=Specified Neural Progenitors Sort to Form Sharp Domains after Noisy Shh Signaling|journal=Cell|language=en|volume=153|issue=3|pages=550–561|doi=10.1016/j.cell.2013.03.023|pmc=3674856|pmid=23622240}}</ref><ref>{{cite journal |last1=Akieda |first1=Yuki |last2=Ogamino |first2=Shohei |last3=Furuie |first3=Hironobu |last4=Ishitani |first4=Shizuka |last5=Akiyoshi |first5=Ryutaro |last6=Nogami |first6=Jumpei |last7=Masuda |first7=Takamasa |last8=Shimizu |first8=Nobuyuki |last9=Ohkawa |first9=Yasuyuki |last10=Ishitani |first10=Tohru |title=Cell competition corrects noisy Wnt morphogen gradients to achieve robust patterning in the zebrafish embryo |journal=Nature Communications |date=17 October 2019 |volume=10 |issue=1 |page=4710 |doi=10.1038/s41467-019-12609-4 |pmid=31624259 |pmc=6797755 |bibcode=2019NatCo..10.4710A }}</ref><ref>{{Cite journal|last1=Kesavan|first1=Gokul|last2=Hans|first2=Stefan|last3=Brand|first3=Michael|date=2019|title=Cell-fate plasticity, adhesion and cell sorting complementarily establish a sharp midbrain-hindbrain boundary|journal=bioRxiv|volume=147|issue=11|language=en|doi=10.1101/857870|pmid=32439756|doi-access=free|url=https://www.biorxiv.org/content/biorxiv/early/2019/11/27/857870.full.pdf}}</ref>

Additionally, the structure (or topology) of [[Signal transduction|signaling pathways]] has been demonstrated to play an important role in robustness to genetic perturbations.<ref>{{Cite journal|last1=Eldar|first1=Avigdor|last2=Rosin|first2=Dalia|last3=Shilo|first3=Ben-Zion|last4=Barkai|first4=Naama|date=2003|title=Self-Enhanced Ligand Degradation Underlies Robustness of Morphogen Gradients|journal=Developmental Cell|language=en|volume=5|issue=4|pages=635–646|doi=10.1016/S1534-5807(03)00292-2|pmid=14536064|doi-access=free}}</ref> Self-enhanced degradation has long been an example of robustness in [[Systems biology|System biology]].<ref>{{cite journal |last1=Ibañes |first1=Marta |last2=Belmonte |first2=Juan Carlos Izpisúa |title=Theoretical and experimental approaches to understand morphogen gradients |journal=Molecular Systems Biology |date=25 March 2008 |volume=4 |issue=1 |page=176 |doi=10.1038/msb.2008.14 |pmid=18364710 |pmc=2290935 }}</ref> Similarly, robustness of dorsoventral patterning in many species emerges from the balanced shuttling-degradation mechanisms involved in [[Bone morphogenetic protein|BMP signaling]].<ref>{{cite journal |last1=Eldar |first1=Avigdor |last2=Dorfman |first2=Ruslan |last3=Weiss |first3=Daniel |last4=Ashe |first4=Hilary |last5=Shilo |first5=Ben-Zion |last6=Barkai |first6=Naama |title=Robustness of the BMP morphogen gradient in Drosophila embryonic patterning |journal=Nature |date=September 2002 |volume=419 |issue=6904 |pages=304–308 |doi=10.1038/nature01061 |pmid=12239569 |bibcode=2002Natur.419..304E |s2cid=4397746 }}</ref><ref>{{Cite journal|last1=Genikhovich|first1=Grigory|last2=Fried|first2=Patrick|last3=Prünster|first3=M. Mandela|last4=Schinko|first4=Johannes B.|last5=Gilles|first5=Anna F.|last6=Fredman|first6=David|last7=Meier|first7=Karin|last8=Iber|first8=Dagmar|last9=Technau|first9=Ulrich|date=2015|title=Axis Patterning by BMPs: Cnidarian Network Reveals Evolutionary Constraints|journal=Cell Reports|language=en|volume=10|issue=10|pages=1646–1654|doi=10.1016/j.celrep.2015.02.035|pmc=4460265|pmid=25772352}}</ref><ref>{{Cite journal|last1=Al Asafen|first1=Hadel|last2=Bandodkar|first2=Prasad U.|last3=Carrell-Noel|first3=Sophia|last4=Reeves|first4=Gregory T.|date=2019-08-19|title=Robustness of the Dorsal morphogen gradient with respect to morphogen dosage|language=en|doi=10.1101/739292|doi-access=free|url=https://www.biorxiv.org/content/biorxiv/early/2019/09/05/739292.full.pdf}}</ref>

==Evolutionary consequences==
Since organisms are constantly exposed to genetic and non-genetic perturbations, robustness is important to ensure the stability of [[phenotypes]].  Also, under mutation-selection balance, mutational robustness can allow [[cryptic genetic variation]] to accumulate in a population.  While phenotypically neutral in a stable environment, these genetic differences can be revealed as trait differences in an environment-dependent manner (see [[evolutionary capacitance]]), thereby allowing for the expression of a greater number of heritable phenotypes in populations exposed to a variable environment.<ref>{{cite journal |doi=10.1016/j.tig.2009.07.005 |author=Masel J Siegal ML |title=Robustness: mechanisms and consequences |journal=Trends in Genetics|volume=25 |issue=9 |pages=395–403 |year=2009 |pmid=19717203 |pmc=2770586}}</ref>

Being robust may even be a favoured at the expense of total fitness as an [[evolutionarily stable strategy]] (also called survival of the flattest).<ref>{{cite journal|last=Wilke|first=CO|author2=Wang, JL |author3=Ofria, C |author4=Lenski, RE |author5= Adami, C |title=Evolution of digital organisms at high mutation rates leads to survival of the flattest.|journal=Nature|date=Jul 19, 2001|volume=412|issue=6844|pages=331–3|pmid=11460163|doi=10.1038/35085569|bibcode=2001Natur.412..331W|s2cid=1482925|url=https://authors.library.caltech.edu/records/br5zq-k4b57/files/412331a0_S1.pdf?download=1}}</ref> A high but narrow peak of a [[fitness landscape]] confers high fitness but low robustness as most mutations lead to massive loss of fitness. High mutation rates may favour population of lower, but broader fitness peaks. More critical biological systems may also have greater selection for robustness as reductions in function are more damaging to [[fitness (biology)|fitness]].<ref name="Van Dijk et al.">{{cite journal | author = Van Dijk | title = Mutational Robustness of Gene Regulatory Networks | journal = PLOS ONE | year = 2012 | volume = 7 | issue = 1 | article-number = e30591 | doi = 10.1371/journal.pone.0030591 | pmid = 22295094 | last2 = Van Mourik | first2 = Simon | last3 = Van Ham | first3 = Roeland C. H. J. | pmc=3266278|display-authors=etal| bibcode = 2012PLoSO...730591V | doi-access = free }} {{open access}}</ref> Mutational robustness is thought to be one driver for theoretical [[viral quasispecies]] formation.[[File:Neutral network.png|thumb|400px|Each circle represents a functional gene variant and lines represent point mutations between them. Light grid-regions have low [[fitness (biology)|fitness]], dark regions have high fitness. ('''a''') White circles have few neutral neighbours, black circles have many. Light grid-regions contain no circles because those sequences have low fitness. ('''b''') Within a neutral network, the population is predicted to evolve towards the centre and away from 'fitness cliffs' (dark arrows).]]

=== Emergent mutational robustness ===
{{see also|Neutral network (evolution)}}
[[Natural selection]] can select directly or indirectly for robustness. When [[mutation rate]]s are high and [[population size]]s are large, populations are predicted to move to more densely connected regions of [[neutral network (evolution)|neutral network]] as less robust variants have fewer surviving mutant descendants.<ref name="van Nimwegen">{{cite journal |  vauthors=van Nimwegen E, Crutchfield JP, Huynen M | title =  Neutral evolution of mutational robustness | journal = PNAS | year = 1999| volume = 96 | issue = 17 | pages = 9716–9720| pmid =  10449760 | pmc =  22276 | doi=10.1073/pnas.96.17.9716| bibcode =1999PNAS...96.9716V| arxiv =adap-org/9903006| doi-access = free }}</ref> The conditions under which selection could act to directly increase mutational robustness in this way are restrictive, and therefore such selection is thought to be limited to only a few [[viruses]]<ref name="Montville">{{cite journal |vauthors=Montville R, Froissart R, Remold SK, Tenaillon O, Turner PE | title =  Evolution of mutational robustness in an RNA virus |  journal = PLOS Biology | year = 2005 | volume = 3| issue = 11| pages = 1939–1945 | doi=10.1371/journal.pbio.0030381| pmid = 16248678 | pmc = 1275523 | doi-access = free }} {{open access}}</ref> and [[microbes]]<ref name="Masel">{{cite journal |  author = Masel J, Maughan H| title =  Mutations Leading to Loss of Sporulation Ability in Bacillus subtilis Are Sufficiently Frequent to Favor Genetic Canalization |  journal = Genetics | year = 2007 | volume = 175| issue =  1 | pages = 453–457 | doi=10.1534/genetics.106.065201| pmid =  17110488 | last2 =  Maughan | pmc =  1775008 }}</ref> having large population sizes and high mutation rates. Such emergent robustness has been observed in experimental evolution of [[cytochrome P450]]s<ref name="Bloom 29">{{cite journal|last=Bloom|first=JD|author2=Lu, Z |author3=Chen, D |author4=Raval, A |author5=Venturelli, OS |author6= Arnold, FH |title=Evolution favors protein mutational robustness in sufficiently large populations.|journal=BMC Biology|date=Jul 17, 2007|volume=5|page=29|pmid=17640347|doi=10.1186/1741-7007-5-29 |pmc=1995189|bibcode=2007arXiv0704.1885B|arxiv=0704.1885 |doi-access=free }} {{open access}}</ref> and [[B-lactamase]].<ref name="Bershtein 2008 1029–1044">{{cite journal|last=Bershtein|first=Shimon|author2=Goldin, Korina |author3=Tawfik, Dan S. |title=Intense Neutral Drifts Yield Robust and Evolvable Consensus Proteins|journal=Journal of Molecular Biology|date=June 2008|volume=379|issue=5|pages=1029–1044|doi=10.1016/j.jmb.2008.04.024|pmid=18495157}}</ref> Conversely, mutational robustness may evolve as a byproduct of natural selection for robustness to environmental perturbations.<ref>{{cite journal |vauthors=Meiklejohn CD, Hartl DL |title=A single mode of canalization|journal=Trends in Ecology & Evolution|volume=17 |issue=10|pages=468–473 |year=2002 |doi=10.1016/s0169-5347(02)02596-x}}</ref><ref>{{cite journal |doi=10.1002/1097-010X(20001015)288:3<242::AID-JEZ5>3.0.CO;2-O |vauthors=Ancel LW, Fontana W |title=Plasticity, evolvability, and modularity in RNA|journal=Journal of Experimental Zoology|volume=288 |issue=3 |pages=242–283 |year=2000 |pmid=11069142|citeseerx=10.1.1.43.6910 }}</ref><ref>{{cite journal |doi=10.1093/molbev/msp008 |vauthors=Szöllősi GJ, Derényi I |title=Congruent Evolution of Genetic and Environmental Robustness in Micro-RNA|journal=Molecular Biology and Evolution|volume=26 |issue=4 |pages=867–874 |year=2009 |pmid=19168567 |arxiv=0810.2658 |s2cid=8935948 }}</ref><ref>{{cite journal |doi=10.2307/2411105 |pmid=28565347 |vauthors=Wagner GP, Booth G, Bagheri-Chaichian H | title=A population genetic theory of canalization | journal=Evolution|volume=51 |issue=2 |pages=329–347|year=1997 |jstor=2411105}}</ref><ref>{{cite journal |author=Lehner B  |title=Genes Confer Similar Robustness to Environmental, Stochastic, and Genetic Perturbations in Yeast|journal=PLOS ONE |volume=5 |issue=2 |pages=468–473 |year=2010 |pmid=20140261 |pmc=2815791 |doi=10.1371/journal.pone.0009035 |bibcode=2010PLoSO...5.9035L|doi-access=free}} {{open access}}</ref>

=== Robustness and evolvability ===
{{see also|Evolvability|Canalisation (genetics)}}
Mutational robustness has been thought to have a negative impact on [[evolvability]] because it reduces the mutational accessibility of distinct heritable phenotypes for a single genotype and reduces selective differences within a genetically diverse population.{{Citation needed|date=March 2014}} Counter-intuitively however, it has been hypothesized that phenotypic robustness towards mutations may actually increase the pace of heritable phenotypic adaptation when viewed over longer periods of time.<ref name=Draghi2010>{{cite journal |doi=10.1038/nature08694 |title=Mutational robustness can facilitate adaptation |year=2010 |last1=Draghi |first1=Jeremy A. |last2=Parsons |first2=Todd L. |last3=Wagner |first3=Günter P. |last4=Plotkin |first4=Joshua B. |journal=Nature |volume=463 |issue=7279 |pages=353–5 |pmid=20090752 |pmc=3071712|bibcode = 2010Natur.463..353D }}</ref><ref name=Wagner2008>{{cite journal |doi=10.1098/rspb.2007.1137 |title=Robustness and evolvability: A paradox resolved |year=2008 |last1=Wagner |first1=A. |journal=Proceedings of the Royal Society B: Biological Sciences |volume=275 |issue=1630 |pages=91–100 |pmid=17971325 |pmc=2562401 |jstor=25249473}}</ref><ref>{{cite journal |doi=10.1016/j.tig.2010.06.002 |vauthors=Masel J, Trotter MV |title=Robustness and evolvability |journal=Trends in Genetics|volume=26 |issue=9 |pages=406–414 |year=2010 |pmid=20598394 |pmc=3198833}}</ref><ref name="Aldana et al.">{{cite journal |  author = Aldana | title =  Robustness and evolvability in genetic regulatory networks |  journal = Journal of Theoretical Biology | year = 2007 | volume = 245 | issue = 3 | pages = 433–448 | pmid =  17188715 | author2 =  Balleza, E | author3 =  Kauffman, S | author4 =  Resendiz, O | doi =  10.1016/j.jtbi.2006.10.027 | bibcode = 2007JThBi.245..433A |display-authors=etal}}</ref>

One hypothesis for how robustness promotes evolvability in asexual populations is that connected networks of fitness-neutral genotypes result in mutational robustness which, while reducing accessibility of new heritable phenotypes over short timescales, over longer time periods, neutral mutation and [[genetic drift]] cause the population to spread out over a larger [[Neutral network (evolution)|neutral network]] in genotype space.<ref>{{cite journal |author1=Ebner, Marc |author2=Shackleton, Mark |author3=Shipman, Rob |title=How neutral networks influence evolvability |journal=Complexity |date=2001 |volume=7 |issue=2 |pages=19–33 |doi=10.1002/cplx.10021 |bibcode=2001Cmplx...7b..19E }}</ref> This genetic diversity gives the population mutational access to a greater number of distinct heritable phenotypes that can be reached from different points of the neutral network.<ref name=Draghi2010 /><ref name=Wagner2008 /><ref name="Aldana et al."/><ref name="Babajide et al.">{{cite journal |  author = Babajide | title =  Neutral networks in protein space: A computational study based on knowledge-based potentials of mean force |  journal = Folding & Design | year = 1997 | volume = 2 | issue = 5 | pages = 261–269 | pmid =  9261065 | author2 =  Hofacker, I. L. | author3 =  Sippl, M. J. | author4 =  Stadler, P. F. | doi=10.1016/s1359-0278(97)00037-0|display-authors=etal| doi-access = free }}</ref><ref name="van Nimwegen and Crutchfield">{{cite journal |  author = van Nimwegen and Crutchfield | title =  Metastable evolutionary dynamics: Crossing fitness barriers or escaping via neutral paths? |  journal = Bulletin of Mathematical Biology | year = 2000 | volume = 62 | issue = 5 | pages = 799–848 | pmid =  11016086 | doi =  10.1006/bulm.2000.0180 | arxiv = adap-org/9907002 | s2cid = 17930325 }}</ref><ref name="Ciliberti et al.">{{cite journal |  author = Ciliberti| title =  Innovation and robustness in complex regulatory gene networks |  journal = Proceedings of the National Academy of Sciences, USA | year = 2007 | volume = 104 | issue = 34 | pages = 13591–13596 | doi=10.1073/pnas.0705396104| pmid = 17690244 | pmc = 1959426 |display-authors=etal|bibcode =2007PNAS..10413591C| doi-access = free }}</ref><ref>{{cite journal |  author = Andreas Wagner | title =  Neutralism and selectionism: a network-based reconciliation |  journal = Nature Reviews Genetics | year = 2008 | volume = 9 | pages=965–974 |doi=10.1038/nrg2473 | issue=12 | pmid=18957969| s2cid = 10651547 | url = https://www.zora.uzh.ch/id/eprint/6535/2/Wagner_NRG_2008.pdf }}</ref> However, this mechanism may be limited to phenotypes dependent on a single genetic locus; for polygenic traits, genetic diversity in asexual populations does not significantly increase evolvability.<ref>{{cite journal|last=Rajon|first=E.|author2=Masel, J. |title=Compensatory Evolution and the Origins of Innovations|journal=Genetics|date=18 January 2013|volume=193|issue=4|pages=1209–1220|doi=10.1534/genetics.112.148627|pmid=23335336|pmc=3606098}}</ref>

In the case of proteins, robustness promotes evolvability in the form of an excess free energy of [[protein folding|folding]].<ref name="Bloom et al.">{{cite journal |  author = Bloom| title =  Protein stability promotes evolvability |  journal = Proceedings of the National Academy of Sciences | year = 2006 | volume = 103 | issue = 15 | pages = 5869–74 | doi=10.1073/pnas.0510098103 | pmid=16581913 | pmc=1458665|display-authors=etal| bibcode=2006PNAS..103.5869B| doi-access = free }}</ref> Since most mutations reduce stability, an excess folding free energy allows toleration of mutations that are beneficial to activity but would otherwise destabilise the protein.

In sexual populations, robustness leads to the accumulation of cryptic genetic variation with high evolutionary potential.<ref>{{cite book|author=Waddington CH |title=The strategy of the genes |year=1957 |publisher=George Allen & Unwin}}</ref><ref>{{cite journal|last=Masel|first=J.|title=Cryptic Genetic Variation Is Enriched for Potential Adaptations|journal=Genetics|date=30 December 2005|volume=172|issue=3|pages=1985–1991|doi=10.1534/genetics.105.051649|pmid=16387877|pmc=1456269}}</ref>

Evolvability may be high when robustness is reversible, with [[evolutionary capacitance]] allowing a switch between high robustness in most circumstances and low robustness at times of stress.<ref>{{cite journal|last=Masel|first=J|title=Q&A: Evolutionary capacitance.|journal=BMC Biology|date=Sep 30, 2013|volume=11|page=103|doi=10.1186/1741-7007-11-103|pmid=24228631|pmc=3849687|doi-access=free}} {{open access}}</ref>

== Methods and model systems ==
There are many systems that have been used to study robustness. ''In silico'' models have been used to model [[Promoter (genetics) | promoters]],<ref name="Bianco-2022">{{cite journal | last=Bianco | first=Simone | title=Artificial Intelligence: Bioengineers' Ultimate Best Friend | journal=[[GEN Biotechnology]] | publisher=[[Mary Ann Liebert]] | volume=1 | issue=2 | date=2022-04-01 | issn=2768-1572 | doi=10.1089/genbio.2022.29027.sbi | pages=140–141 | s2cid=248313305}}</ref><ref>{{cite journal | vauthors = Vaishnav ED, de Boer CG, Molinet J, Yassour M, Fan L, Adiconis X, Thompson DA, Levine JZ, Cubillos FA, Regev A  | title = The evolution, evolvability and engineering of gene regulatory DNA | journal = Nature | volume = 603 | issue = 7901 | pages = 455–463 | date = March 2022 | pmid = 35264797 | doi = 10.1038/s41586-022-04506-6| pmc = 8934302 | bibcode = 2022Natur.603..455V }}</ref> [[RNA secondary structure]], [[protein lattice model]]s, or [[Wagner's gene network model|gene networks]]. Experimental systems for individual genes include enzyme activity of [[cytochrome P450]],<ref name="Bloom 29"/> [[B-lactamase]],<ref name="Bershtein 2008 1029–1044"/> [[RNA polymerase]],<ref name="Guo 9205–10"/> and [[LacI]]<ref name="Guo 9205–10"/> have all been used. Whole organism robustness has been investigated in [[RNA virus]] fitness,<ref name="Sanjuán 1975–82"/> bacterial [[chemotaxis]], ''[[Drosophila]]'' fitness,<ref name="Baba 2006 2006.0008"/> segment polarity network, neurogenic network and [[bone morphogenetic protein]] gradient, ''[[Caenorhabditis elegans|C. elegans]]'' fitness<ref name="Davies 1748–1751"/> and [[vulva]]l development, and mammalian [[circadian clock]].<ref name=Wagner2005>{{cite book |author=Wagner A |title=Robustness and evolvability in living systems |publisher = Princeton University Press |series=Princeton Studies in Complexity |year=2005 |isbn=0-691-12240-7}}{{page needed|date=July 2013}}</ref>

== See also ==
* [[Distribution of fitness effects]]
* [[Evolvability]]
* [[Canalisation (genetics)|Canalization]]
* [[Neutral network (evolution)]]
* [[Epistasis]]
* [[Evolutionary capacitance]]
* [[Fitness landscape]]
* [[Evolutionary developmental biology]]

== References ==
{{reflist|35em}}
{{genarch}}

{{DEFAULTSORT:Robustness (Evolution)}}
[[Category:Evolutionary biology]]
[[Category:Genetics]]