Convergent pathways of reductive mitochondrial evolution characterised with hypercubic inference

Convergent pathways of reductive mitochondrial evolution characterised with hypercubic inference | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Convergent pathways of reductive mitochondrial evolution characterised with hypercubic inference View ORCID Profile Robert C. Glastad , View ORCID Profile Iain G. Johnston doi: https://doi.org/10.1101/2025.04.29.651065 Robert C. Glastad 1 Department of Mathematics, University of Bergen , Norway Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Robert C. Glastad Iain G. Johnston 1 Department of Mathematics, University of Bergen , Norway 2 Computational Biology Unit, University of Bergen , Norway Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Iain G. Johnston For correspondence: iain.johnston{at}uib.no Abstract Full Text Info/History Metrics Data/Code Preview PDF Abstract For a striking example of mitochondrial behaviour beyond ATP generation, consider mitochondrion-related organelles (MROs). Hydrogenosomes, mitosomes, and other reduced mitochondrial forms have evolved through the loss of physical and functional features, from individual ETC complexes to oxidative phosphorylaytion and the very ability to produce ATP (and further). Reduction of mitochondria is a dramatic example of convergent evolution, occuring in every eukaryotic kingdom and many parallel times. Here, we use hypercubic inference, a class of methods from evolutionary accumulation modelling (EvAM), to explore the pathways of convergent mitochondrial reduction across eukaryotes. We find that most MRO diversity can be explained by small variations on two distinct pathways, starting with either the loss of Complex I or the loss of Complexes III/IV, which tend to proceed over different characteristic timescales. We show that different clades, including ciliates and apicomplexans, reflect particular instances of these pathways. Using metabolic modelling, we connect the structure of these evolutionary pathways to the metabolic impact of the changes involved, suggesting a plausible explanation for the dramatically convergent nature of reductive evolution. We discuss this approach in connection with related theory on the genetic and functional reduction of mitochondria across organisms. Introduction Mitochondria are central, multifunctional organelles in eukaryotic cells ( Monzel et al., 2023 ). Originally independent organisms, over evolutionary history they have become progressively entrained by their “host” cells, with most of their genes having been lost or transferred to the host nucleus ( Giannakis et al., 2022 ; Janouškovec et al., 2017 ; Johnston & Williams, 2016 ; Roger et al., 2017 ; Smith & Keeling, 2015 ; Veeraragavan et al., 2024 ). In some species – including well-known cases in yeasts ( Hackstein et al., 2006 ), parasitic plants ( Petersen et al., 2020 ; Senkler et al., 2018 ), and anaerobic protists ( Leger et al., 2019 ; Makiuchi & Nozaki, 2014; Mathur et al., 2021 ) – this genetic reduction has been accompanied by the loss of physical and functional features of mitochondria ( Maciszewski & Karnkowska, 2019 ; Makiuchi & Nozaki, 2014 ; Stairs et al., 2015 ). These features include electron transport chain (ETC) protein complexes, transporters, the mitochondrial genome, elements of the tricarboxylic acid (TCA) cycle, and even the ability to produce ATP. The collection of species with observed mitochondrial feature losses continues to expand, including discoveries of the first eukaryote lacking mitochondria ( Karnkowska et al., 2016 ), the first animal lacking mtDNA ( Yahalomi et al., 2020 ), and increasing characterisation of reduced mitochondria in protists ( Mathur et al., 2021 ; Záhonová et al., 2023 ). These highly reduced organelles can generally be referred to as mitochondrion-related organelles or MROs ( de Paula et al., 2012 ; Hampl & Roger, 2024 ; Leger et al., 2019 ; Lewis et al., 2020 ; Makiuchi & Nozaki, 2014; Van Der Giezen, 2009 ). They include hydrogenosomes (which retain the ability to produce ATP, including through a characteristic pathway producing molecular hydrogen), mitosomes (which are highly reduced and do not produce ATP) and other categories (sometimes labelled class I-V dependent on metabolic status ( Müller et al., 2012 )). Reduced mitochondria and MROs retain metabolic importance in their carrier species (notably, machinery involved in iron-sulfur cluster assembly seems to be near-universally retained in MROs), with some remaining features postulated as targets for therapeutic interventions ( Goodman et al., 2017 ). But the loss of canonical mitochondrial features often leads to unusual behaviour in these MROs. In the absence of a complete proton-pumping electron transport chain (ETC), ATP synthase may run in reverse, consuming ATP to maintain mitochondrial membrane potential. Genes for mitochondrial machinery may be retained without lab-detectable activity of that machinery ( Gerasimov et al., 2025 ; Verner et al., 2011 ). Loss of proton-pumping ETC complexes may be partly compensated by non-proton-pumping alternatives, including alternative NADH dehydrogenases and alternative oxidases (Fang & Beattie, 2003; Mathur et al., 2021 ; Rasmusson et al., 2008 ). Clearly, these reductions highlight roles of mitochondria(−related organelles) beyond ATP production. MROs have been found in diverse and deeply-branching eukaryotic lineages, making the loss of mitochondrial features a striking example of convergent evolution ( Lewis et al., 2020 ; Mathur et al., 2021 ; Stairs et al., 2015 ; Van Der Giezen, 2009 ). As such, similarities and differences in the “pathways” of the loss process across different clades – the orderings and timings with which features are lost – are points of fundamental biological interest. Algorithmic developments in the study of convergent evolutionary processes have made it possible to quantify structure, variability, and uncertainty in the pathways by which such processes take place ( Aga et al., 2024 ; Beerenwinkel et al., 2005 ; Boyko & Beaulieu, 2021 , 2023; Diaz-Uriarte & Herrera-Nieto, 2022 ; Renz et al., 2024 ; Williams et al., 2013 ). While these approaches have been applied to organelle genome evolution ( Giannakis et al., 2022 ; Giannakis, Aga, et al., 2024 ; Johnston & Williams, 2016 ), they have not to our knowledge yet been applied to this class of reductive mitochondrial evolution. Here, we aim to apply these “hypercubic inference” approaches – a class of methods within evolutionary accumulation modelling or EvAM ( Diaz-Uriarte & Herrera-Nieto, 2022 ; Diaz-Uriarte & Johnston, 2024 ) -- to describe the possible evolutionary pathways of mitochondrial feature loss across eukaryotes. Methods Dataset compilation We required a set of features that could (when required) be estimated from genetic information, that have previously been well studied, that reflect different aspects of mitochondrial functionality, and which could be captured by simple (and therefore generalisable) metabolic modelling approaches. Motivated by an extensive literature on the key features categorising mitochondrial reduction, the features we consider (with shorthand labels) are: the TCA cycle ( TCA ), ETC respiratory protein complexes I-IV ( CI ; CII ; CIII ; CIV ), ATP synthase ( CV ), the mtDNA genome ( DNA ), pyruvate dehydrogenase ( PDH ), and machinery required for iron-sulfur cluster production ( Fe-S ). We took advantage of the extensive literature on the topic to compile lists from meta-analysis of MROs as well as a collection of individual species where these features were characterised (reference list in Supp. Text 1). The data, with sources and discussion, is available at the Github repository below. Binarisation To assign binary presence-absence markers to these sometimes nuanced features, we make several feature-specific choices. For ETC complexes, we consider “presence” to mean the canonical versions of complexes found in most eukaryotes. Some yeasts and protists have replaced canonical CI with an alternative NADH dehydrogenase; this counts as loss in our protocol. Some organisms have replaced CIV (and CIII) with alternative oxidases as terminal electron acceptors and/or replaced PDH with a non-canonical alternative ( Mathur et al., 2021 ); we also count these as losses. For the TCA cycle, we consider the complete, common TCA cycle as “presence” and incomplete cycles as “absence”. Tree construction We used both NCBI’s Common Taxonomy Tree ( Federhen, 2012 ) to obtain a coarse-grained taxonomy linking species, and TimeTree ( Kumar et al., 2022 ) for the subset of available species for a smaller phylogeny with estimated timings. Hypercubic transition path sampling (HyperTraPS) HyperTraPS considers how multiple, potentially coupled, binary features coevolve on a phylogeny ( Aga et al., 2024 ; Greenbury et al., 2020 ; Johnston & Williams, 2016 ). Given a phylogeny where each tip has a collection of L binary features (for example, Fig. 1A ), the approach first reconstructs ancestral states through the phylogeny. Here, we assume that mitochondrial feature losses are relatively rare and irreversible, so that an ancestor is inferred to have lost a feature if and only if all its descendants have lost it (otherwise, loss is assumed to occur independently in the subset of descendant lineages lacking the feature). Each ancestor-descendant pair on the phylogeny then constitutes an observed “before-after” pair of states (which may be identical), and a timescale (if the phylogeny has meaningful branch lengths). Download figure Open in new tab Figure 1. Data and inference for pathways of reductive mitochondrial evolution across eukaryotes. (A) Source data, comprising the taxonomy of species and the presence (grey) or absence (white) of each mitochondrial feature in each species. Selected clades are labelled. (B) “Bubble plot” from HyperTraPS inference. The area of each circle gives the probability with which the corresponding feature (rows) is lost at the corresponding ordering (columns) in a reductive evolution pathway. For example, Fe-S is overwhelmingly likely to be the last feature lost; CI has two peaks in loss probability, as the first or eigth feature to be lost. (C) The most likely transitions through the evolutionary space inferred by HyperTraPS. Each edge (line segment) corresponds to the loss of a feature, with width giving the probability of that loss; each node (point between edges) corresponds to a particular set of presences and absences. Nodes corresponding to some example species are labelled. The topmost state is the ancestral state with all features present; the bottommost state has no features remaining; reductive evolution proceeds down the figure. For example, the state at the far left of the figure has lost CI, CIII , and TCA , and its most likely next loss is CIV . HyperTraPS works by considering a “state space” of all the 2 L possible binary states that could, in principle, be observed in an evolutionary system. Transitions through this space are allowed from a “before” state to an “after” state that differs from the before state by the loss of exactly one feature. All possible paths from the state 111… (retaining all features) to the state 000… (having lost all features) are therefore possible. Each transition is parameterised by a probability, or a rate (in the continuous time case). HyperTraPS then identifies parameters for this space of evolutionary transitions that are most compatible with the set of “before-after” pairs of states observed in the dataset. By default (and here), this is a Bayesian inference process using Markov chain Monte Carlo ( Aga et al., 2024 ). After checking for numerical convergence, the results can be summarised in various ways, including the probability that a given feature is the n th to be lost in the reduction process ( Fig. 1B ) or a summary of the specific inferred transition parameters in the evolutionary space ( Fig. 1C ). By default (and here), HyperTraPS considers parameters describing (a) the base rate of loss of each feature and (b) how this base rate is affected by the loss of each other feature (Johnston & Williams, 2016) – a scheme also known as mutual hazard networks ( Schill et al., 2020 ). In this way, interactions between feature losses – whether the loss of A makes the loss of B more or less likely – can also be quantified. Flux balance analysis We used MitoMAMMAL, a recent genome-scale metabolic model of mammalian mitochondria, as our model for non-reduced mitochondrial metabolism ( Chapman et al., 2025 ). Using sybil ( Gelius-Dietrich et al., 2013 ) and sybilSBML ( Fritzemeier et al., 2017 ) packages in R, we performed flux balance analysis ( Orth et al., 2010 ) to explore influences of reaction losses on mitochondrial(−related organelle) metabolism. We set either ATP production or generation of proton-motive force as the objective function of interest, and modelled the influence of removing reactions corresponding to ETC Complexes I-V, individual TCA cycle steps, and pyrvuate dehydrogenase. We modelled metabolism both under default oxygen conditions and under hypoxic conditions, modelled with oxygen levels decreased to 10% of the default case. To model the presence of compensatory, non-canonical ETC subunits, we introduced reactions: AOX: 2 ubiquinol + O 2 → 2 ubiquinone + H 2 O NDHalt: H + + NADH + 0.999 ubiquinone + 0.002 O 2 → NAD + 0.999 ubiquinol + 0.002 O 2 · - These two reactions serve as an electron acceptor and an NADH reducer respectively without pumping protons. The small deviations from integers in the stoichiometry of the NDHalt reaction are taken from the original MitoMAMMAL CI reaction, and reflect some small amount of superoxide production. Superoxide production has also been suggested in alternative NADH dehydrogenases in yeast (Fang & Beattie, 2003), but the presence or absence of superoxide is not an consequential term in our metabolic analysis and will not influence our findings. Software In addition to the packages referenced above, we used R ( R Core Team, 2022 ) with ggplot2 ( Wickham, 2016 ), ggpubr ( Kassambara, 2020 ), ggraph ( Pedersen, 2020 ), ggtree and ggtreeExtra ( Xu et al., 2021 ; Yu et al., 2017 ) for visualisation; igraph ( Csardi & Nepusz, 2006 ) for network analysis; and ape ( Paradis & Schliep, 2019 ), phangorn ( Schliep, 2011 ), and phytools ( Revell, 2012 ) for phylogenetics. The specific parameters used for the HyperTraPS algorithm were 200 sampling walkers for likelihood calculations, 10 5 MCMC steps, no likelihood penalisation, and a normal perturbation kernel with standard deviation 0.5, checking for convergence across three numerical replicates ( Aga et al., 2024 ). Our pipeline and dataset is freely available at https://github.com/StochasticBiology/MROs . Results Distinct pathways of mitochondrial feature loss in reductive evolution across eukaryotes We first sought to compile a dataset of patterns of lost mitochondrial features across eukaryotes (Supp. Data 1). This process involved (a) selecting the mitochondrial and MRO features that we would characterise; (b) determining a protocol for labelling each feature “present” or “absent” in each species; (c) compiling a list of species and their presence-absence patterns; and (d) determining the relationship between these species. Details of these steps are given in the Methods. To summarise, we consider losses of canonical ETC complexes ( CI - CV ), loss of any component of the canonical TCA cycle ( TCA ), loss of mtDNA ( DNA ), loss of canonical pyruvate dehydrogenase ( PDH ), and loss of iron-sulfur metabolism capacity ( Fe-S ), across a set of eukaryotes connected by an estimated taxonomy ( Fig. 1A ) and a subset with a better time-resolved phylogeny ( Supp. Fig. 1A ). If a lost feature is replaced by a non-canonical alternative, we still consider it lost. The datasets are shown in Fig. 1A and Supp. Fig. 1A . It should immediately be pointed out that the features we consider are not independent. As CII is a component of the complete TCA cycle, we cannot have loss of CII without loss of TCA . As essential subunits of CIII and CIV are encoded in mtDNA, we cannot have loss of DNA without loss of CIII and CIV – although our dataset contains one exception, Selenidium serpulae , where a mitochondrial genome appears to be absent and CIII-IV genes are speculated to have been transferred to the nucleus ( Mathur et al., 2021 ). Our first step in analysing these data was to use HyperTraPS (hypercubic transition path sampling ( Aga et al., 2024 ; Greenbury et al., 2020 ; Johnston & Williams, 2016 )) to infer the likely orderings of feature loss events across these trees of species (see Methods). Briefly, HyperTraPS constructs an evolutionary model where all combinations of feature presence and absence are possible, then asks which weightings of transitions through this space (from a putative ancestor with all features, towards a putative eventual “end state” with no features) are most compatible with the observed data (controlling for relatedness, in the sense that descendants of a common ancestor do not constitute independent representatives of a given set of features (Maddison & FitzJohn, 2015; Revell, 2010 )). First results from this inference are shown in Fig. 1B-C and Supp. Fig. 1B-C . In Fig. 1B and Supp. Fig. 1B , the probability that a given feature is lost at a given step in reductive evolution (for example, the first feature to be lost, the second, and so on) is given. As a “control”, iron-sulfur metabolism ( Fe-S ) is always inferred to be most likely lost last. Complex I ( CI ) has a clearly bimodal loss distributions: it may be lost early, or late, in the reductive process, but less likely at intermediate stages. This immediately suggests multiple distinct evolutionary pathways: one type with early loss of CI and one distinct type with late loss of CI . The presence of multiple distinct pathways is supported by a more detailed look at the inferred transitions through the evolutionary model ( Fig. 1C , Supp. Fig. 1C ). Here, the most likely transitions from the (top) ancestral state towards the (bottom) putative final state are shown. It can immediately be seen that the most likely evolutionary pathways are highly canalised: following a collection of possible first loss events, subsequent evolution follows distinct pathways before converging on Fe-S as the final loss event. Of course, an important issue in any such analysis pipeline is the robustness of results to errors or nuances in the original dataset. We first note that the discrete-time and continuous-time inference processes for the different data subsets in Fig. 1 and Supp. Fig. 1 give compatible results. We also performed a control study where we artificially perturbed the source data by flipping a random 10% of the presence-absence markers for ten resampled datasets (see Methods). The resulting inferred dynamics were largely consistent with each other and with the unperturbed results ( Supp. Fig. 2 ), suggesting that these findings are robust to (at least) this error rate in the source data. Download figure Open in new tab Figure 2. Influences between features, and evolutionary pathway structures. (A) “Motif plots” from HyperTraPS output describing the probability (section height) that each feature is lost at each ordering step in a pathway (the same concept as Fig. 1B ). (i) All pathways, reproducing Fig. 1B . (ii) CI-class pathway, where the first loss is constrained to be CI and the conditional probabilities of following losses are plotted through the pathway. (iii) alt-class pathway, where the first loss is constrained to be either CIII, CIV , or TCA . (B) Inferred influences between feature losses; edge widths given posterior means (in effectively arbitrary units) for parameters with coefficients of variation under 0.65. Blue edges are “activating” ( CIV → CIII means that CIV loss makes CIII loss more likely ) ; red edges are “repressing” ( CIV loss makes CI loss less likely). Inferred pathway structures and influences in reductive mitochondrial evolution HyperTraPS makes it possible to dissect the structure of evolutionary pathways, and the possible interactions between events that support them, in more depth. In Fig. 2A , we plot the inferred probabilities with which different subsequent loss events follow a particular, given first loss event. Strikingly distinct pathways result when this given first loss event is either CI or an alternative. These initial steps lead to rather tightly-defined following pathways inferred to occur with high probability, which we will refer to as: CI-class: following CI, CIII , ( CIV and TCA ), CV, DNA, CII, PDH , (then Fe-S ). alt-class: following ( CIII, CIV , and TCA ), CV, DNA, PDH, CII, CI , (then Fe-S ). It will immediately be seen that the combination of these distinct CI- and alt-class pathways can give rise to the bimodality in CI ordering distribution observed in Fig. 1B . There is also a lower-probability set of pathways involving CI- and alt-class orderings with PDH loss evolving earlier, rather than later. HyperTraPS allows features to influence each other, in the sense that the loss of one feature can have a positive (activating, →) or negative (repressing, ⊣) effect on the probability that another feature is lost. Every feature has a base rate of loss, and the loss of other features can influence this base rate. Fig. 2B shows the collection of these influences that were robustly inferred across eukaryotic species. It is readily seen that positive influences CIV → CIII and TCA → CIV support the observed close links between the first steps of the alt-class pathway, and CI → CIII supports these steps immediately after the CI-class pathway begins. The CIV ⊣ CI interaction supports the distinction between the CI-class and alt-class pathways, ensuring that once the alt-class pathway starts, CI loss is forced later. Mutual repressions between CV and PDH support the bimodal distribution of PDH in the CI-class pathway. A recent approach using hypercubic directed acyclic graphs (HyperDAGs) (Giannakis, Aga, et al., 2024 ) allows a minimal set of evolutionary pathways to be computed for a dataset. In a sense, the most deterministic, reproducible evolutionary system corresponds to a single, canalised evolutionary pathway in Fig. 1C . Every independent lineage would then evolve through the same sequence of loss events. Every instance where one species has A but not B and another species has B but not A (referred to as an “incompatibility”) requires departure from such a single, canalised evolutionary pathway. Such departures take the form of branches – out-degrees greater than 1 – in the evolutionary spaces like Fig. 1C . HyperDAGs finds the smallest number of branches required to account for the incompatibilities in a dataset; here only 6 branches are required to account for all 82 observations ( Supp. Fig. 3 ). Of these, 5 are associated with different early orderings of CIII, CIV , and TCA loss (and are therefore in a sense contained within the alt-class pathway), and the final branch corresponds to whether PDH is acquired early or late when CI is the first loss. Download figure Open in new tab Figure 3. Evolutionary pathways inferred within specific protist phyla. (A) (i) Bubble plot (as in Fig. 1B ) and (ii) transition network (as in Fig. 1C ) for HyperTraPS inference using only the ciliate phylum as input data. (B) (i) Bubble plot and (ii) transition network using only the apicomplexan phylum as input data. Inset images are public domain illustrations from phylopic.org. This suggestion of distinct pathways is supported by individual species profiles in Fig. 1A , including (for example): the well-known mistletoe ( Viscum album ) and yeast ( Saccharomyces cerevisiae ) reflecting early CI-class, Nematosida parisii and Plagiopyla frontata reflecting late CI-class, Brevimastigomonas motovehiculus and liver fluke ( Fasciola hepatica ) reflecting early alt-class, and Neocallimastix sp. and Chilomastix cuspidata reflecting late alt-class. Taken together, these results consistently suggest two distinct, canalised evolutionary pathways of reductive mitochondrial evolution. Clade-specific pathways of inferred reductive mitochondrial evolution Following these observations, we next asked whether these structures in evolutionary pathways across eukaryotes arose from different behaviour in different clades. The first clade we examined was ciliates, a phylum of widespread protists ( Fig. 1A , ( Rotterová et al., 2020 )). Here, while reduction of CI is observed via the loss of subunits, complete loss of the complex seems less common. Accordingly, HyperTraPS shows no support for the CI-class pathway in ciliates, with the alt-class pathway dominating the inferred dynamics ( Fig. 3A-B ). We also explored dynamics in apicomplexans, a phylum of largely parasitic protists ( Fig. 1A , ( Mathur et al., 2021 )). Here, CI is commonly lost, and PDH is also lost across the phylum ( Fig. 3C-D ). Accordingly, HyperTraPS recovers strong support for the CI-class pathway, with the change that PDH is lost as the first or second step rather than later. The inferred evolutionary trajectory in this phylum is very highly canalised ( Fig. 3D ). HyperDAG analysis shows that only one incompatibility exists ( CII vs CV ) so that, in principle, a single evolutionary pathway with one branch can explain all apicomplexan observations. Inference about timescales of reductive evolution pathways For the subset of species with available estimates of divergence times ( Supp. Fig. 1A ), a version of the algorithm called HyperTraPS-CT (continuous time) allows inference about the possible timescales of evolutionary events (as opposed to only considering their relative orderings, as in Figs. 1 - 2 . We exploit HyperTraPS-CT’s capacity to account for uncertainty in timings and assume a time window for each branch between half and twice the point estimate from TimeTree ( Kumar et al., 2022 ). As above, the summaries of results from this continuous time inference ( Supp. Fig. 1B-C ) are compatible with the ordering-only case, with some shifts. First, ordering probabilities are more uniform, reflecting the increased uncertainty from the necessarily smaller dataset. The alt-class and CI-class pathways are roughly equal in probability, and PDH is more likely acquired early in the CI-class pathway. The inferred timescales of loss events are illustrated in Supp. Fig. 4 . The distributions of likely timings of loss events are very broad – both owing to the diversity of lineages where the losses occur and the relatively small dataset size. Despite this breadth, a general trend is visible, where the distributions of alt-class loss steps are shifted towards shorter timescales than for CI-class (though with considerable overlap). This reflects the set of species where CI is lost with no other losses (resembling a “stalling” of the loss process) versus the set of species where multiple losses in the CIII, CIV, TCA group, and followup steps in the alt-class pathway, have occurred in comparatively rapid succession. Notably, the first three steps of the alt-class pathway and the second to fourth steps of the CI-class pathway – in both cases most likely to involve CIII, CIV, TCA losses – occur within what is comparatively a very short timescale of each other. Download figure Open in new tab Figure 4. Flux balance analysis of the metabolic impact of feature loss on ATP production. The computationally predicted metabolic impact of the loss of single features and sets of features in our study, using MitoMAMMAL ( Chapman et al., 2025 ). Each column reflects either the non-reduced case (“None”), a single computational knockout of a metabolic reaction, or a set of simultaneous knockouts. The height of a bar gives predicted ATP production under that knockout experiment. Grey circles give the predicted ATP production when alternative oxidase is not modelled as a terminal electron acceptor (see Methods). Individual TCA cycle reactions are labelled with shorthand acronyms; impactful reactions are citrate synthase (CS) and malate dehydrogenase (MDH). Other conditions and objectives are shown in Supp. Fig. 5 . Connecting inferred loss dynamics and simulated metabolic impact of lost mitochondrial features We next asked what the metabolic implications of these different patterns of mitochondrial feature loss could be. Working as generally as possible, we used MitoMAMMAL, a recent genome-scale model of mammalian mitochondrial metabolism ( Chapman et al., 2025 ) (reflecting an ancestral, non-reduced organelle). We performed a constraint-based computational analysis of the metabolic impact of each loss on objective functions of ATP production and proton-motive force generation in normoxic and hypoxic conditions, including in the presence of introduced reactions reflecting non-canonical replacements of ETC complexes (see Methods). Under normoxic conditions ( Fig. 4 ), from the ETC complexes, the loss of CI has the lowest effect on ATP production, followed by the loss of CIII and/or CIV with compensatory gain of AOX. Without compensatory gain of AOX, CIII/IV loss immediately compromises ETC activity due to the lack of a terminal electron acceptor, reducing ATP production to its minimal level (no OXPHOS activity). PDH loss, either alone or in conjunction with other lost features, has a very limited impact on ATP production. Complete loss of the TCA cycle substantially compromises ATP production, but losses of many single steps (which would constitute TCA loss in our evolutionary analysis) have very little effect on ATP production; citrate synthase, CII, and malate dehydrogenase are exceptions where ATP production is more substantially reduced. The simulation under hypoxic conditions, and with proton-motive force rather than ATP generation as the metabolic objective have similar orderings of CI, CIII, CIV , and TCA features, with only differences in the relative importance of CV for the proton-motive force objective and CII for the hypoxic conditions ( Supp. Fig. 5 ). Once CI and any one of CIII, CIV, TCA are lost, ATP production approaches a minimum level. Hence, from an ATP production perspective, there is no advantage in retaining the other two members of the set. This compatible with the inference that loss of these features proceed in rapid succession in the CI-class pathway. Once either CIII or CIV is lost, loss of the other (or loss of many TCA steps) provides no further ATP production impact, again supporting the idea that this group of features can be lost in a highly coupled way. Taking these observations together, there is a clear correlation between the modelled impact of feature loss on ATP production and the propensity of that feature to be lost early in our inferred evolutionary pathways. As such, for circumstances where ATP production remains an important component of fitness, the CI- and alt-class pathways may then reflect the smoothest routes of progress through intermediate states on a fitness landscape (see Discussion). Discussion We have shown that the phylogenetically diverse, parallel instances of reductive mitochondrial evolution can be explained by a relatively limited set of canalised evolutionary pathways – and that in turn, these pathways can in part be explained by the relative metabolic impacts of the different feature losses involved. In particular, mitochondrial features seem to be lost roughly according to the consequent impact on metabolism. A connection between reductive evolution of mtDNA gene content and metabolic demands from organismal environment and ecology has been proposed ( García-Pascual et al., 2022 ; Giannakis, Richards, et al., 2024; Veeraragavan et al., 2024 ). Broadly, organelles in organisms with metabolism influenced by strong environmental oscillations (day/night cycles, tidal cycles) are predicted to retain more genes locally, while organelles in more stable and/or less demanding environments are freer to transfer genes to the nucleus. This idea borrows strongly from the theory of colocation for redox regulation (CoRR), suggesting that retaining organelle genes supports local, rapid, individual control of organelle machinery ( Allen, 2015 ; Allen & Martin, 2016; Allen & Raven, 1996). In this picture, reduced pressure to retain local control of mitochondrial function relaxes the pressure to retain mitochondrial genes in mtDNA, and favours their relocation to the safer nucleus. While many other factors are likely involved in shaping the reduction of mtDNA and mitochondria across species ( Butenko et al., 2024 ; Johnston & Williams, 2016; Lynch et al., 2006 ; Smith, 2016 ), the loss of the features we consider here can in a sense be viewed as an extension of this picture, where the function of some mitochondrial machinery is under so little retention pressure that it is lost entirely. The properties of MROs have indeed been cited as evidence for CoRR over competing hypotheses for gene retention ( de Paula et al., 2012 ). Evolutionary accumulation modelling (EvAM) is at the core of our analysis here. EvAM approaches have often been applied only to cross-sectional observations, largely to study cancer progression in independent patients ( Diaz-Uriarte & Johnston, 2024 ). EvAM approaches supporting phylogenetically-embedded data ( Aga et al., 2024 ; Greenbury et al., 2020 ; Johnston & Williams, 2016 ; Luo et al., 2023 ; Moen & Johnston, 2023) have started to bridge the gap with more established tools from phylogenetic comparative methods (Boyko & Beaulieu, 2021; O’Meara, 2012 ; Pagel, 1994 ), with recent work explicitly phrasing EvAM as a well-known Mk model (Johnston & Diaz-Uriarte, 2024) and many EvAM approaches falling into such a continuous-time Markov chain with finite state space (CTMC-FSS) picture ( O’Meara, 2012 ). We hope that this study illustrates that EvAM can provide insights into evolutionary systems involving the coupled evolution of multiple binary traits, complementing its existing use cases from C 4 photosynthesis ( Williams et al., 2013 ) to tool use in animals (Johnston & Røyrvik, 2020). Our compilation and analysis of data from diverse eukaryotes is subject to several potential issues. Even the primary source data for the presence and absence of mitochondrial features is subject to some uncertainty (for example, when a feature is lost but its genes remain, or when a feature remains but is not detectable in particular experiments); the choices made in our binarisation of that source data may also be questioned. Species names and taxonomic positioning may also be updated over time – for example, since beginning this manuscript we have learned of a proposed reclassification for Sawyeria marylandensis into genus Psalteriomonas ( Foučková et al., 2024 ). Some concerns about the robustness of our findings to data errors and changes may be resolved by the fact that introducing random errors in the source data did not qualitatively change the outcomes ( Supp. Fig. 2 ), but we would always appreciate corrections for any obvious errors in our compilation. Further, in our effort to be general we have omitted specific details and features. The diversity of compensatory, non-canonical features gained by reduced mitochondria is not captured in our coarse-grained metabolic modelling; we do not consider transporters (like AAC), import machinery, the presence of other metabolic pathways, or other interesting MRO features in our dataset (Makiuchi & Nozaki, 2014). A clear target for future work is to expand this feature set, building on careful characterisation of these features across eukaryotes, which will increase the detail with which these convergence reduction pathways can be described. Supplementary Information Download figure Open in new tab Supplementary Figure 1. HyperTraPS-CT inference using continuous time. (A) Source data, with branch lengths now derived from divergence time estimates with TimeTree ( Kumar et al., 2022 ). For scale, the top two species ( T. brucei and L. tropica ) have a divergence time of 266 Ma. (B) Bubble plot and (C) transition network from HyperTraPS-CT inference, now using a possible time window for each transition of width [0.5 τ , 1.5 τ ] where τ is the estimated divergence time. Timescale distributions for these pathways are shown in Supp. Fig. 4 . Download figure Open in new tab Supplementary Figure 2. Robustness of inferred pathways with respect to uncertainty in source data. Bubble plot from HyperTraPS inference. Each segment represents the inference from a resampled dataset, where 10% of the data are artificially changed from 0 to 1 or vice versa. Overall the inferred dynamics are robust to these sets of perturbations, with segments showing the same trends across resamples. Download figure Open in new tab Supplementary Figure 3. Hypercubic directed acyclic graph (HyperDAG) analysis of minimal pathways. A set of edges on the transition network that can account for all observed states in the source data ( Fig. 1A ) and display minimal branching – keeping as close as possible to a single, deterministic evolutionary pathway (Giannakis, Aga, et al., 2024 ). Each node is labelled by a binary string describing the features at which losses have (1) or have not (0) occurred: the ordering of features in these strings is CI, CII, CIII, CIV, CV, PDH, TCA, DNA, TCA, Fe-S . The top state, as in Fig. 1C , is the ancestral state with no losses. There is then a set of branches to CI (1000…), CIII, CIV, TCA first losses. The branches from CIV first loss (00010…) account for different patterns of the early alt-class steps. The branch from CI first loss accounts for either CIII or PDH being the next loss. Apart from these branches, all other states can be accounted for by canalised pathways, which descend the remainder of the evolutionary space. Download figure Open in new tab Supplementary Figure 4. Inferred timescales of loss events in CI- and alt-class pathways. HyperTraPS-CT infers the distribution of timescales for each loss event in a model evolutionary process. The bars here give (mean ± standard deviation) for (log) timings of each step in a set of simulated CI-class and alt-class pathways on the fitted hypercubic model. For example, the first step for CI has a range of timings of 150-1900 Ma. Download figure Open in new tab Supplementary Figure 5. Flux balance analysis of mitochondrial feature loss under different conditions. (A) Normoxic and (B) hypoxic (10% normoxic) conditions simulated with simulated knockouts (KO) of different individual and paired mitochondrial features, as in Fig. 4 . Red bars give flux values for ATP production under an objective of maximising ATP production, blue bars give proton motive force (PMF) under an objective of maximising PMF. PMF values are scaled by 1/3.7 to align with ATP values in the default model case with no knockouts. Supplementary Text 1. Additional references used in the meta-analysis of mitochondrial feature profiles . Numbers correspond to labels in the annotated dataset. Acknowledgments We are very grateful to Hassan Hashimi, Vladimír Hampl, Tomáš Pánek, Ingrid Sveráková, and other delegates of the 54th Jírovec’s Protozoological Days conference for feedback on the data and inferences in this study, and to Ellen Røyrvik, Paul Una, Amelia Earl, and Ai Monti for valuable discussions. This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 805046 (EvoConBiO) to IGJ). This work was supported by the Trond Mohn Foundation [project HyperEvol under grant agreement No. TMS2021TMT09 to IGJ], through the Centre for Antimicrobial Resistance in Western Norway (CAMRIA) [TMS2020TMT11]. Funder Information Declared European Research Council, https://ror.org/0472cxd90 , 805046 Trond Mohn Stiftelse, , TMS2021TMT09 , TMS2020TMT11 Footnotes https://github.com/StochasticBiology/MROs References ↵ Aga , O. N. , Brun , M. , Giannakis , K. , Dauda , K. A. , Diaz-Uriarte , R. , & Johnston , I. G. ( 2024 ). HyperTraPS-CT: Inference and prediction for accumulation pathways with flexible data and model structures . PLOS Computational Biology , 20 ( 9 ), e1012393 . OpenUrl CrossRef ↵ Allen , J. F. ( 2015 ). Why chloroplasts and mitochondria retain their own genomes and genetic systems: Colocation for redox regulation of gene expression . Proceedings of the National Academy of Sciences , 112 ( 33 ), 10231 – 10238 . doi: 10.1073/pnas.1500012112 OpenUrl Abstract / FREE Full Text Allen , J. F. , & Martin , W. F. ( 2016 ). Why Have Organelles Retained Genomes? Cell Systems , 2 ( 2 ), 70 – 72 . doi: 10.1016/j.cels.2016.02.007 OpenUrl CrossRef PubMed Allen , J. F. , & Raven , J. A. ( 1996 ). Free-radical-induced mutation vs redox regulation: Costs and benefits of genes in organelles . Journal of Molecular Evolution , 42 , 482 – 492 . doi: 10.1007/BF02352278 OpenUrl CrossRef PubMed Web of Science ↵ Beerenwinkel , N. , Däumer , M. , Sing , T. , Rahnenführer , J. , Lengauer , T. , Selbig , J. , HoVmann , D. , & Kaiser , R. ( 2005 ). Estimating HIV evolutionary pathways and the genetic barrier to drug resistance . The Journal of Infectious Diseases , 191 ( 11 ), 1953 – 1960 . OpenUrl CrossRef PubMed Web of Science ↵ Boyko , J. D. , & Beaulieu , J. M. ( 2021 ). Generalized hidden Markov models for phylogenetic comparative datasets . Methods in Ecology and Evolution , 12 ( 3 ), 468 – 478 . doi: 10.1111/2041-210X.13534 OpenUrl CrossRef ↵ Boyko , J. D. , & Beaulieu , J. M. ( 2023 ). Reducing the biases in false correlations between discrete characters . Systematic Biology , 72 ( 2 ), 476 – 488 . OpenUrl CrossRef PubMed ↵ Butenko , A. , Lukeš , J. , Speijer , D. , & Wideman , J. G. ( 2024 ). Mitochondrial genomes revisited: Why do diVerent lineages retain diVerent genes? BMC Biology , 22 ( 1 ), 15 . doi: 10.1186/s12915-024-01824-1 OpenUrl CrossRef PubMed ↵ Chapman , S. , Brunet , T. , Mourier , A. , & Habermann , B. H. ( 2025 ). MitoMAMMAL: A genome scale model of mammalian mitochondria predicts cardiac and BAT metabolism . Bioinformatics Advances , 5 ( 1 ), vbae172 . doi: 10.1093/bioadv/vbae172 OpenUrl CrossRef ↵ Csardi , G. , & Nepusz , T. ( 2006 ). The igraph software package for complex network research . InterJournal, Complex Systems , 1695 ( 5 ), 1 – 9 . OpenUrl CrossRef ↵ C. E. Bullerwell de Paula , W. B. M. , Allen , J. F. , & van der Giezen , M. ( 2012 ). Mitochondria, Hydrogenosomes and Mitosomes in Relation to the CoRR Hypothesis for Genome Function and Evolution . In C. E. Bullerwell (Ed.), Organelle Genetics: Evolution of Organelle Genomes and Gene Expression (pp. 105 – 119 ). Springer . doi: 10.1007/978-3-642-22380-8_5 OpenUrl CrossRef ↵ Diaz-Uriarte , R. , & Herrera-Nieto , P. ( 2022 ). EvAM-Tools: Tools for evolutionary accumulation and cancer progression models . Bioinformatics , 38 ( 24 ), 5457 – 5459 . OpenUrl CrossRef PubMed Diaz-Uriarte , R. , & Johnston , I. G. ( 2024 ). A picture guide to cancer progression and monotonic accumulation models: Evolutionary assumptions, plausible interpretations, and alternative uses ( arXiv: 2312.06824 ). arXiv . doi: 10.48550/arXiv.2312.06824 OpenUrl CrossRef Fang , J. , & Beattie , D. S. ( 2003 ). External alternative NADH dehydrogenase of Saccharomyces cerevisiae: A potential source of superoxide . Free Radical Biology and Medicine , 34 ( 4 ), 478 – 488 . doi: 10.1016/S0891-5849(02)01328-X OpenUrl CrossRef PubMed ↵ Federhen , S. ( 2012 ). The NCBI Taxonomy database . Nucleic Acids Research , 40 ( D1 ), D136 – D143 . doi: 10.1093/nar/gkr1178 OpenUrl CrossRef PubMed Web of Science ↵ Foučková , M. , Uhrová , K. , Kubánková , A. , Pánek , T. , & Č epička , I. ( 2024 ). Lighting lantern above Psalteriomonadidae: Unveiling novel diversity within the genus Psalteriomonas (Discoba: Heterolobosea) . European Journal of Protistology , 93 , 126052 . doi: 10.1016/j.ejop.2024.126052 OpenUrl CrossRef PubMed ↵ Fritzemeier , C. J. , Gelius-Dietrich , G. , & Alzoubi , D. ( 2017 ). Habil, A. sybilSBML: SBML Integration in Package ‘Sybil’ . ↵ García-Pascual , B. , Nordbotten , J. M. , & Johnston , I. G. ( 2022 ). Cellular and environmental dynamics influence species-specific extents of organelle gene retention . Proceedings of the Royal Society B: Biological Sciences , 290 ( 1994 ), 20222140 . doi: 10.1098/rspb.2022.2140 OpenUrl CrossRef ↵ Gelius-Dietrich , G. , Desouki , A. A. , Fritzemeier , C. J. , & Lercher , M. J. ( 2013 ). Sybil – EVicient constraint-based modelling in R . BMC Systems Biology , 7 ( 1 ), 125 . doi: 10.1186/1752-0509-7-125 OpenUrl CrossRef PubMed ↵ Gerasimov , E. S. , Afonin , D. A. , Škodová-Sveráková , I. , Saura , A. , Trusina , N. , Gahura , O. , Zakharova , A. , Butenko , A. , Baráth , P. , Horváth , A. , Opperdoes , F. R. , Pérez-Morga , D. , Zimmer , S. L. , Lukeš , J. , & Yurchenko , V. ( 2025 ). Evolutionary divergent kinetoplast genome structure and RNA editing patterns in the trypanosomatid Vickermania . Proceedings of the National Academy of Sciences , 122 ( 15 ), e2426887122 . doi: 10.1073/pnas.2426887122 OpenUrl CrossRef PubMed Giannakis , K. , Aga , O. N. L. , Moen , M. T. , Drange , P. G. , & Johnston , I. G. ( 2024 ). Identifying parsimonious pathways of accumulation and convergent evolution from binary data (p. 2024.11.06.622201) . bioRxiv . doi: 10.1101/2024.11.06.622201 OpenUrl Abstract / FREE Full Text ↵ Giannakis , K. , Arrowsmith , S. J. , Richards , L. , Gasparini , S. , Chustecki , J. M. , Røyrvik , E. C. , & Johnston , I. G. ( 2022 ). Evolutionary inference across eukaryotes identifies universal features shaping organelle gene retention . Cell Systems , 13 ( 11 ), 874 - 884.e5 . doi: 10.1016/j.cels.2022.08.007 OpenUrl CrossRef PubMed Giannakis , K. , Richards , L. , & Johnston , I. G. ( 2024 ). Ecological predictors of organelle genome evolution: Phylogenetic correlations with taxonomically broad, sparse, unsystematized data . Systematic Biology, syae009 . doi: 10.1093/sysbio/syae009 OpenUrl CrossRef ↵ Goodman , C. D. , Buchanan , H. D. , & McFadden , G. I. ( 2017 ). Is the Mitochondrion a Good Malaria Drug Target? Trends in Parasitology , 33 ( 3 ), 185 – 193 . doi: 10.1016/j.pt.2016.10.002 OpenUrl CrossRef PubMed ↵ Greenbury , S. F. , Barahona , M. , & Johnston , I. G. ( 2020 ). HyperTraPS: Inferring probabilistic patterns of trait acquisition in evolutionary and disease progression pathways . Cell Systems , 10 ( 1 ), 39 – 51 . OpenUrl CrossRef PubMed ↵ Hackstein , J. H. P. , Tjaden , J. , & Huynen , M. ( 2006 ). Mitochondria, hydrogenosomes and mitosomes: Products of evolutionary tinkering! Current Genetics , 50 ( 4 ), 225 – 245 . doi: 10.1007/s00294-006-0088-8 OpenUrl CrossRef PubMed Web of Science ↵ S. D. Schwartzbach , P. G. Kroth , & M. Oborník Hampl , V. , & Roger , A. J. ( 2024 ). The Evolutionary Origin of Mitochondria and Mitochondrion-Related Organelles . In S. D. Schwartzbach , P. G. Kroth , & M. Oborník (Eds.), Endosymbiotic Organelle Acquisition: Solutions to the Problem of Protein Localization and Membrane Passage (pp. 89 – 121 ). Springer International Publishing . doi: 10.1007/978-3-031-57446-7_3 OpenUrl CrossRef ↵ Janouškovec , J. , Tikhonenkov , D. V. , Burki , F. , Howe , A. T. , Rohwer , F. L. , Mylnikov , A. P. , & Keeling , P. J. ( 2017 ). A New Lineage of Eukaryotes Illuminates Early Mitochondrial Genome Reduction . Current Biology , 27 ( 23 ), 3717 - 3724.e5 . doi: 10.1016/j.cub.2017.10.051 OpenUrl CrossRef PubMed Johnston , I. G. , & Diaz-Uriarte , R. ( 2024 ). A hypercubic Mk model framework for capturing reversibility in disease, cancer, and evolutionary accumulation modelling (p. 2024.06.27.600959) . bioRxiv . doi: 10.1101/2024.06.27.600959 OpenUrl Abstract / FREE Full Text Johnston , I. G. , & Røyrvik , E. C. ( 2020 ). Data-driven inference reveals distinct and conserved dynamic pathways of tool use emergence across animal taxa . Iscience , 23 ( 6 ). ↵ Johnston , I. G. , & Williams , B. P. ( 2016 ). Evolutionary Inference across Eukaryotes Identifies Specific Pressures Favoring Mitochondrial Gene Retention . Cell Systems , 2 ( 2 ), 101 – 111 . doi: 10.1016/j.cels.2016.01.013 OpenUrl CrossRef PubMed ↵ Karnkowska , A. , Vacek , V. , Zubáčová , Z. , Treitli , S. C. , Petrželková , R. , Eme , L. , Novák , L. , Žársk? , V. , Barlow , L. D. , & Herman , E. K. ( 2016 ). A eukaryote without a mitochondrial organelle . Current Biology , 26 ( 10 ), 1274 – 1284 . doi: 10.1016/j.cub.2016.03.053 OpenUrl CrossRef PubMed ↵ Kassambara , A. ( 2020 ). Ggpubr:”ggplot2” based publication ready plots . R Package Version 0.4. 0 , 438 . ↵ Kumar , S. , Suleski , M. , Craig , J. M. , Kasprowicz , A. E. , Sanderford , M. , Li , M. , Stecher , G. , & Hedges , S. B. ( 2022 ). TimeTree 5: An Expanded Resource for Species Divergence Times . Molecular Biology and Evolution , 39 ( 8 ), msac174 . doi: 10.1093/molbev/msac174 OpenUrl CrossRef PubMed ↵ J. Tachezy Leger , M. M. , Kolísko , M. , Stairs , C. W. , & Simpson , A. G. B. ( 2019 ). Mitochondrion-Related Organelles in Free-Living Protists . In J. Tachezy (Ed.), Hydrogenosomes and Mitosomes: Mitochondria of Anaerobic Eukaryotes (pp. 287 – 308 ). Springer International Publishing . doi: 10.1007/978-3-030-17941-0_12 OpenUrl CrossRef ↵ Lewis , W. H. , Lind , A. E. , Sendra , K. M. , Onsbring , H. , Williams , T. A. , Esteban , G. F. , Hirt , R. P. , Ettema , T. J. G. , & Embley , T. M. ( 2020 ). Convergent Evolution of Hydrogenosomes from Mitochondria by Gene Transfer and Loss . Molecular Biology and Evolution , 37 ( 2 ), 524 – 539 . doi: 10.1093/molbev/msz239 OpenUrl CrossRef PubMed ↵ Luo , X. G. , Kuipers , J. , & Beerenwinkel , N. ( 2023 ). Joint inference of exclusivity patterns and recurrent trajectories from tumor mutation trees . Nature Communications , 14 ( 1 ), Article 1. doi: 10.1038/s41467-023-39400-w OpenUrl CrossRef PubMed ↵ Lynch , M. , Koskella , B. , & Schaack , S. ( 2006 ). Mutation Pressure and the Evolution of Organelle Genomic Architecture . Science , 311 ( 5768 ), 1727 – 1730 . doi: 10.1126/science.1118884 OpenUrl Abstract / FREE Full Text ↵ Maciszewski , K. , & Karnkowska , A. ( 2019 ). Should I stay or should I go? Retention and loss of components in vestigial endosymbiotic organelles . Current Opinion in Genetics & Development , 58–59 , 33 – 39 . doi: 10.1016/j.gde.2019.07.013 OpenUrl CrossRef Maddison , W. P. , & FitzJohn , R. G. ( 2015 ). The Unsolved Challenge to Phylogenetic Correlation Tests for Categorical Characters . Systematic Biology , 64 ( 1 ), 127 – 136 . doi: 10.1093/sysbio/syu070 OpenUrl CrossRef PubMed ↵ Makiuchi , T. , & Nozaki , T. ( 2014 ). Highly divergent mitochondrion-related organelles in anaerobic parasitic protozoa . Biochimie , 100 , 3 – 17 . doi: 10.1016/j.biochi.2013.11.018 OpenUrl CrossRef PubMed ↵ Mathur , V. , Wakeman , K. C. , & Keeling , P. J. ( 2021 ). Parallel functional reduction in the mitochondria of apicomplexan parasites . Current Biology , 31 ( 13 ), 2920 – 2928 . doi: 10.1016/j.cub.2021.04.028 OpenUrl CrossRef Moen , M. T. , & Johnston , I. G. ( 2023 ). HyperHMM: EVicient inference of evolutionary and progressive dynamics on hypercubic transition graphs . Bioinformatics , 39 ( 1 ), btac803 . OpenUrl CrossRef PubMed ↵ Monzel , A. S. , Enríquez , J. A. , & Picard , M. ( 2023 ). Multifaceted mitochondria: Moving mitochondrial science beyond function and dysfunction . Nature Metabolism , 5 ( 4 ), 546 – 562 . doi: 10.1038/s42255-023-00783-1 OpenUrl CrossRef PubMed ↵ Müller , M. , Mentel , M. , van Hellemond , J. J. , Henze , K. , Woehle , C. , Gould , S. B. , Yu , R.-Y. , van der Giezen , M. , Tielens , A. G. M. , & Martin , W. F. ( 2012 ). Biochemistry and Evolution of Anaerobic Energy Metabolism in Eukaryotes . Microbiology and Molecular Biology Reviews , 76 ( 2 ), 444 – 495 . doi: 10.1128/mmbr.05024-11 OpenUrl Abstract / FREE Full Text ↵ O’Meara , B. C. ( 2012 ). Evolutionary Inferences from Phylogenies: A Review of Methods . Annual Review of Ecology, Evolution, and Systematics , 43 (Volume 43, 2012), 267–285. doi: 10.1146/annurev-ecolsys-110411-160331 OpenUrl CrossRef ↵ Orth , J. D. , Thiele , I. , & Palsson , B.Ø. ( 2010 ). What is flux balance analysis? Nature Biotechnology , 28 ( 3 ), Article 3. doi: 10.1038/nbt.1614 OpenUrl CrossRef PubMed Web of Science ↵ Pagel , M. ( 1994 ). Detecting correlated evolution on phylogenies: A general method for the comparative analysis of discrete characters . Proceedings of the Royal Society of London. Series B: Biological Sciences , 255 ( 1342 ), 37 – 45 . OpenUrl CrossRef ↵ Paradis , E. , & Schliep , K. ( 2019 ). ape 5.0: An environment for modern phylogenetics and evolutionary analyses in R . Bioinformatics , 35 ( 3 ), 526 – 528 . doi: 10.1093/bioinformatics/bty633 OpenUrl CrossRef PubMed ↵ Pedersen , T. L. ( 2020 ). ggraph: An implementation of grammar of graphics for graphs and networks . R Package Version , 2 ( 3 ), 1 . OpenUrl ↵ Petersen , G. , Anderson , B. , Braun , H.-P. , Meyer , E. H. , & Møller , I. M. ( 2020 ). Mitochondria in parasitic plants . Mitochondrion , 52 , 173 – 182 . doi: 10.1016/j.mito.2020.03.008 OpenUrl CrossRef PubMed ↵ R Core Team . ( 2022 ). R: A language and environment for statistical computing. R Foundation for Statistical Computing , Vienna, Austria. 2012 . ↵ Rasmusson , A. G. , Geisler , D. A. , & Møller , I. M. ( 2008 ). The multiplicity of dehydrogenases in the electron transport chain of plant mitochondria . Mitochondrion , 8 ( 1 ), 47 – 60 . doi: 10.1016/j.mito.2007.10.004 OpenUrl CrossRef PubMed Web of Science ↵ Renz , J. , Dauda , K. A. , Aga , O. N. L. , Diaz-Uriarte , R. , Löhr , I. H. , Blomberg , B. , & Johnston , I. G. ( 2024 ). Evolutionary accumulation modelling in AMR: Machine learning to infer and predict evolutionary dynamics of multi-drug resistance ( arXiv: 2411.00219 ). arXiv . doi: 10.48550/arXiv.2411.00219 OpenUrl CrossRef ↵ Revell , L. J. ( 2010 ). Phylogenetic signal and linear regression on species data . Methods in Ecology and Evolution , 1 ( 4 ), 319 – 329 . doi: 10.1111/j.2041-210X.2010.00044.x OpenUrl CrossRef ↵ Revell , L. J. ( 2012 ). phytools: An R package for phylogenetic comparative biology (and other things) . Methods in Ecology and Evolution , 2 , 217 – 223 . OpenUrl ↵ Roger , A. J. , Muñoz-Gómez , S. A. , & Kamikawa , R. ( 2017 ). The Origin and Diversification of Mitochondria . Current Biology: CB , 27 ( 21 ), R1177 – R1192 . doi: 10.1016/j.cub.2017.09.015 OpenUrl CrossRef PubMed ↵ Rotterová , J. , Salomaki , E. , Pánek , T. , Bourland , W. , Žihala , D. , Táborsk? , P. , Edgcomb , V. P. , Beinart , R. A. , Kolísko , M. , & Čepička , I. ( 2020 ). Genomics of new ciliate lineages provides insight into the evolution of obligate anaerobiosis . Current Biology , 30 ( 11 ), 2037 – 2050 . OpenUrl CrossRef PubMed ↵ Schill , R. , Solbrig , S. , Wettig , T. , & Spang , R. ( 2020 ). Modelling cancer progression using mutual hazard networks . Bioinformatics , 36 ( 1 ), 241 – 249 . OpenUrl CrossRef PubMed ↵ Schliep , K. P. ( 2011 ). phangorn: Phylogenetic analysis in R . Bioinformatics , 27 ( 4 ), 592 – 593 . doi: 10.1093/bioinformatics/btq706 OpenUrl CrossRef PubMed Web of Science ↵ Senkler , J. , Rugen , N. , Eubel , H. , Hegermann , J. , & Braun , H.-P. ( 2018 ). Absence of Complex I Implicates Rearrangement of the Respiratory Chain in European Mistletoe . Current Biology , 28 ( 10 ), 1606 - 1613.e4 . doi: 10.1016/j.cub.2018.03.050 OpenUrl CrossRef PubMed ↵ Smith , D. R. ( 2016 ). The mutational hazard hypothesis of organelle genome evolution: 10 years on . Molecular Ecology , 25 ( 16 ), 3769 – 3775 . doi: 10.1111/mec.13742 OpenUrl CrossRef Smith , D. R. , & Keeling , P. J. ( 2015 ). Mitochondrial and plastid genome architecture: Reoccurring themes, but significant diVerences at the extremes . Proceedings of the National Academy of Sciences , 112 ( 33 ), 10177 – 10184 . doi: 10.1073/pnas.1422049112 OpenUrl Abstract / FREE Full Text ↵ Stairs , C. W. , Leger , M. M. , & Roger , A. J. ( 2015 ). Diversity and origins of anaerobic metabolism in mitochondria and related organelles . Philosophical Transactions of the Royal Society B: Biological Sciences , 370 ( 1678 ), 20140326 . doi: 10.1098/rstb.2014.0326 OpenUrl CrossRef PubMed ↵ Van Der Giezen , M. ( 2009 ). Hydrogenosomes and Mitosomes: Conservation and Evolution of Functions . Journal of Eukaryotic Microbiology , 56 ( 3 ), 221 – 231 . doi: 10.1111/j.1550-7408.2009.00407.x OpenUrl CrossRef PubMed Web of Science ↵ Veeraragavan , S. , Johansen , M. , & Johnston , I. G. ( 2024 ). Evolution and maintenance of mtDNA gene content across eukaryotes . Biochemical Journal , 481 ( 15 ), 1015 – 1042 . doi: 10.1042/BCJ20230415 OpenUrl CrossRef PubMed ↵ Verner , Z. , Č ermáková , P. , Škodová , I. , Kriegová , E. , Horváth , A. , & Lukeš , J. ( 2011 ). Complex I (NADH:ubiquinone oxidoreductase) is active in but non-essential for procyclic Trypanosoma brucei . Molecular and Biochemical Parasitology , 175 ( 2 ), 196 – 200 . doi: 10.1016/j.molbiopara.2010.11.003 OpenUrl CrossRef PubMed ↵ Wickham , H. ( 2016 ). ggplot2: Elegant Graphics for Data Analysis . Springer-Verlag . https://link.springer.com/book/10.1007/978-0-387-98141-3 ↵ Williams , B. P. , Johnston , I. G. , CovshoV , S. , & Hibberd , J. M. ( 2013 ). Phenotypic landscape inference reveals multiple evolutionary paths to C4 photosynthesis . eLife , 2 , e00961 . doi: 10.7554/eLife.00961 OpenUrl CrossRef PubMed ↵ Xu , S. , Dai , Z. , Guo , P. , Fu , X. , Liu , S. , Zhou , L. , Tang , W. , Feng , T. , Chen , M. , Zhan , L. , Wu , T. , Hu , E. , Jiang , Y. , Bo , X. , & Yu , G. ( 2021 ). ggtreeExtra: Compact Visualization of Richly Annotated Phylogenetic Data . Molecular Biology and Evolution , 38 ( 9 ), 4039 – 4042 . doi: 10.1093/molbev/msab166 OpenUrl CrossRef PubMed ↵ Yahalomi , D. , Atkinson , S. D. , Neuhof , M. , Chang , E. S. , Philippe , H. , Cartwright , P. , Bartholomew , J. L. , & Huchon , D. ( 2020 ). A cnidarian parasite of salmon (Myxozoa: Henneguya) lacks a mitochondrial genome . Proceedings of the National Academy of Sciences , 117 ( 10 ), 5358 – 5363 . doi: 10.1073/pnas.1909907117 OpenUrl Abstract / FREE Full Text ↵ Yu , G. , Smith , D. K. , Zhu , H. , Guan , Y. , & Lam , T. T.-Y. ( 2017 ). ggtree: An r package for visualization and annotation of phylogenetic trees with their covariates and other associated data . Methods in Ecology and Evolution , 8 ( 1 ), 28 – 36 . doi: 10.1111/2041-210X.12628 OpenUrl CrossRef PubMed ↵ Záhonová , K. , Füssy , Z. , Stairs , C. W. , Leger , M. M. , Tachezy , J. , Č epička , I. , Roger , A. J. , & Hampl , V. ( 2023 ). Comparative analysis of mitochondrion-related organelles in anaerobic amoebozoans . Microbial Genomics , 9 ( 11 ), 001143 . doi: 10.1099/mgen.0.001143 OpenUrl CrossRef PubMed 1. Roger , A. J. , Muñoz-Gómez , S. A. , & Kamikawa , R. ( 2017 ). The origin and diversification of mitochondria . Current Biology , 27 ( 21 ), R1177 – R1192 . OpenUrl CrossRef PubMed 2. Maguire , F. , & Richards , T. A. ( 2014 ). Organelle evolution: a mosaic of ‘mitochondrial’functions . Current Biology , 24 ( 11 ), R518 – R520 . OpenUrl CrossRef PubMed 3. J. Hackstein , J. Tjaden , and M. Huynen . Mitochondria, hydrogenosomes and mitosomes: products of evolutionary tinkering! Current Genetics , 50 : 225 , 2006 . OpenUrl CrossRef PubMed Web of Science 4. Stairs , C. W. , Leger , M. M. , & Roger , A. J. ( 2015 ). Diversity and origins of anaerobic metabolism in mitochondria and related organelles . Philosophical Transactions of the Royal Society B: Biological Sciences , 370 ( 1678 ), 20140326 . OpenUrl CrossRef PubMed 5. via NCBI 6. Petersen , G. , Anderson , B. , Braun , H. P. , Meyer , E. H. , & Møller , I. M. ( 2020 ). Mitochondria in parasitic plants . Mitochondrion , 52 , 173 – 182 . OpenUrl CrossRef PubMed 7. Xi , Zhenxiang , et al. “ Massive mitochondrial gene transfer in a parasitic flowering plant clade .” PLoS Genetics 9 . 2 ( 2013 ): e1003265 . OpenUrl CrossRef 8. Gatica-Soria , Leonardo M. , et al. “ Native and foreign mitochondrial and nuclear encoded proteins conform the OXPHOS complexes of a holoparasitic plant .” Gene 817 ( 2022 ): 146176 . OpenUrl CrossRef PubMed 9. Yahalomi , Dayana , et al. “ A cnidarian parasite of salmon (Myxozoa: Henneguya) lacks a mitochondrial genome .” Proceedings of the National Academy of Sciences 117 . 10 ( 2020 ): 5358 – 5363 . OpenUrl Abstract / FREE Full Text 10. Braymer , Joseph J. , et al. “ Mechanistic concepts of iron-sulfur protein biogenesis in Biology .” Biochimica et Biophysica Acta (BBA)-Molecular Cell Research 1868 . 1 ( 2021 ): 118863 . OpenUrl CrossRef 11. Fan , W. , Zhu , A. , Kozaczek , M. et al. Limited mitogenomic degradation in response to a parasitic lifestyle in Orobanchaceae . Sci Rep 6 , 36285 ( 2016 ). doi: 10.1038/srep36285 OpenUrl CrossRef PubMed 12. Zervas , A. , Petersen , G. & Seberg , O. Mitochondrial genome evolution in parasitic plants . BMC Evol Biol 19 , 87 ( 2019 ). doi: 10.1186/s12862-019-1401-8 OpenUrl CrossRef PubMed 13. Sanchez-Puerta , M. Virginia , et al. “ Unparalleled replacement of native mitochondrial genes by foreign homologs in a holoparasitic plant .” New Phytologist 214 . 1 ( 2017 ): 376 – 387 . OpenUrl CrossRef PubMed 14. Tasbihi , Minoo , et al. “ Mitochondrial proteome profiling of Leishmania tropica .” Microbial Pathogenesis 133 ( 2019 ): 103542 . OpenUrl CrossRef PubMed 15. Sandberg , T.O.M. , Yahalomi , D. , Bracha , N. et al. Evolution of myxozoan mitochondrial genomes: insights from myxobolids . BMC Genomics 25 , 388 ( 2024 ). doi: 10.1186/s12864-024-10254-w OpenUrl CrossRef PubMed 16. Yahalomi , Dayana , et al. “ The multipartite mitochondrial genome of Enteromyxum leei (Myxozoa): eight fast-evolving megacircles .” Molecular biology and evolution 34 . 7 ( 2017 ): 1551 – 1556 . OpenUrl CrossRef PubMed 17. Lonergan , Kim M. , and Michael W. Gray . “ Editing of transfer RNAs in Acanthamoeba castellanii mitochondria .” Science 259 . 5096 ( 1993 ): 812 – 816 . OpenUrl Abstract / FREE Full Text 18. Lavrov , Dennis V. , and Walker Pett . “ Animal mitochondrial DNA as we do not know it: mt-genome organization and evolution in nonbilaterian lineages .” Genome biology and evolution 8 . 9 ( 2016 ): 2896 – 2913 . OpenUrl CrossRef PubMed 19. Muthye , Viraj R. , et al. “ The highly repetitive genome of Myxobolus rasmusseni, an emerging myxozoan parasite of fathead minnows .” Genome Biology and Evolution 16 . 11 ( 2024 ): evae220 . OpenUrl CrossRef PubMed 20. Takeuchi , Fumihiko , et al. “ The mitochondrial genomes of a myxozoan genus Kudoa are extremely divergent in Metazoa .” PLoS One 10 . 7 ( 2015 ): e0132030 . OpenUrl CrossRef PubMed 21. John , Uwe , et al. “ An aerobic eukaryotic parasite with functional mitochondria that likely lacks a mitochondrial genome .” Science Advances 5 . 4 ( 2019 ): eaav1110 . OpenUrl FREE Full Text 22. Qing , Xiong , et al. “ Mitochondrial genome of Brugia malayi microfilariae isolated from a clinical sample .” Frontiers in Ecology and Evolution 9 ( 2021 ): 637805 . OpenUrl CrossRef 23. Williams , S. A. , et al. “ The filarial genome project: analysis of the nuclear, mitochondrial and endosymbiont genomes of Brugia malayi .” International journal for parasitology 30 . 4 ( 2000 ): 411 – 419 . OpenUrl CrossRef PubMed Web of Science 24. Keddie , Elise M. , Tarig Higazi , and Thomas R. Unnasch . “ The mitochondrial genome of Onchocerca volvulus: sequence, structure and phylogenetic analysis .” Molecular and biochemical parasitology 95 . 1 ( 1998 ): 111 – 127 . OpenUrl CrossRef PubMed Web of Science 25. Otero , Lucía , et al. “ Complex I and II subunit gene duplications provide increased fitness to worms .” Frontiers in Genetics 10 ( 2019 ): 1043 . OpenUrl CrossRef PubMed 26. Bay Ömer F. , et al. “ A genome-scale metabolic model of parasitic whipworm .” Nature Communications 14 . 1 ( 2023 ): 6937 . OpenUrl CrossRef PubMed 27. Doyle , Stephen R. , et al. “ Population genomics of ancient and modern Trichuris trichiura .” Nature communications 13 . 1 ( 2022 ): 3888 . OpenUrl CrossRef PubMed 28. Shao , Renfu , Ewen F. Kirkness , and Stephen C. Barker . “ The single mitochondrial chromosome typical of animals has evolved into 18 minichromosomes in the human body louse, Pediculus humanus .” Genome research 19 . 5 ( 2009 ): 904 – 912 . OpenUrl Abstract / FREE Full Text 29. Crainey , James L. , et al. “ The mitogenome of Onchocerca volvulus from the Brazilian Amazonia focus .” Memórias do Instituto Oswaldo Cruz 111 . 1 ( 2016 ): 79 – 81 . OpenUrl CrossRef PubMed 30. Ramesh , Akshaya , et al. “ The complete mitochondrial genome sequence of the filarial nematode Wuchereria bancrofti from three geographic isolates provides evidence of complex demographic history .” Molecular and biochemical parasitology 183 . 1 ( 2012 ): 32 – 41 . OpenUrl CrossRef PubMed 31. Makiuchi , Takashi , and Tomoyoshi Nozaki . “ Highly divergent mitochondrion-related organelles in anaerobic parasitic protozoa .” Biochimie 100 ( 2014 ): 3 – 17 . OpenUrl CrossRef PubMed 32. Bu , Xia-lian , et al. “ Mitochondrial metabolism of the facultative parasite Chilodonella uncinata (Alveolata, Ciliophora) .” Parasites & Vectors 16 . 1 ( 2023 ): 92 . OpenUrl CrossRef PubMed 33. Rotterová , Johana , et al. “ Anaerobic ciliates as a model group for studying symbioses in oxygen-depleted environments .” Journal of Eukaryotic Microbiology 69 . 5 ( 2022 ): e12912 . OpenUrl CrossRef PubMed 34. Lewis , William H. , et al. “ Convergent evolution of hydrogenosomes from mitochondria by gene transfer and loss .” Molecular biology and evolution 37 . 2 ( 2020 ): 524 – 539 . OpenUrl CrossRef PubMed 35. Gawryluk , Ryan MR , et al. “ The earliest stages of mitochondrial adaptation to low oxygen revealed in a novel rhizarian .” Current Biology 26 . 20 ( 2016 ): 2729 – 2738 . OpenUrl CrossRef PubMed 36. Chen , Zhicheng , et al. “ Group-specific functional patterns of mitochondrion-related organelles shed light on their multiple transitions from mitochondria in ciliated protists .” Marine Life Science & Technology 4 . 4 ( 2022 ): 609 – 623 . OpenUrl CrossRef PubMed 37. Rotterová , Johana , et al. “ Genomics of new ciliate lineages provides insight into the evolution of obligate anaerobiosis .” Current Biology 30 . 11 ( 2020 ): 2037 – 2050 . OpenUrl CrossRef PubMed 38. Keeling , Patrick J. , et al. “ Parallel functional reduction in the mitochondria of apicomplexan parasites .” European Journal of Protistology 94 ( 2024 ): 126065 . OpenUrl CrossRef PubMed 39. Mathur , Varsha , Kevin C. Wakeman , and Patrick J. Keeling . “ Parallel functional reduction in the mitochondria of apicomplexan parasites .” Current Biology 31 . 13 ( 2021 ): 2920 – 2928 . OpenUrl CrossRef PubMed 40. Leger , Michelle M. , et al. “ Mitochondrion-related organelles in free-living protists .” Hydrogenosomes and mitosomes: mitochondria of anaerobic eukaryotes ( 2019 ): 287 – 308 . 41. Rüdinger , Mareike , et al. “ Plant-type mitochondrial RNA editing in the protist Naegleria gruberi .” Rna 17 . 12 ( 2011 ): 2058 – 2062 . OpenUrl Abstract / FREE Full Text 42. Zhang , Tengteng , et al. “ Characterization and comparative analyses of mitochondrial genomes in single-celled eukaryotes to shed light on the diversity and evolution of linear molecular architecture .” International Journal of Molecular Sciences 22 . 5 ( 2021 ): 2546 . OpenUrl CrossRef PubMed 43. Gawryluk , Ryan MR , and Courtney W. Stairs . “ Diversity of electron transport chains in anaerobic protists .” Biochimica et Biophysica Acta (BBA)-Bioenergetics 1862 . 1 ( 2021 ): 148334 . OpenUrl CrossRef 44. Kayal , Ehsan , and David R. Smith . “ Is the dinoflagellate Amoebophrya really missing an mtDNA? .” Molecular Biology and Evolution 38 . 6 ( 2021 ): 2493 – 2496 . OpenUrl CrossRef PubMed 45. DeMontigny , Wesley , and Tsvetan Bachvarov . “ The Nuclear and Mitochondrial Genomes of Amoebophrya sp. ex Karlodinium veneficum .” G3: Genes, Genomes, Genetics ( 2025 ): jkaf030 . 46. Oborník , Miroslav , and Julius Lukeš. “ The organellar genomes of Chromera and Vitrella, the phototrophic relatives of apicomplexan parasites .” Annual Review of Microbiology 69 . 1 ( 2015 ): 129 – 144 . OpenUrl CrossRef PubMed 47. Vacek , Vojtěch , et al. “ Fe–S cluster assembly in oxymonads and related protists .” Molecular Biology and Evolution 35 . 11 ( 2018 ): 48. Flegontov , Pavel , et al. “ Divergent mitochondrial respiratory chains in phototrophic relatives of apicomplexan parasites .” Molecular biology and evolution 32 . 5 ( 2015 ): 1115 – 1131 . OpenUrl CrossRef PubMed 49. Senkler , Jennifer , et al. “ Absence of complex I implicates rearrangement of the respiratory chain in European mistletoe .” Current Biology 28 . 10 ( 2018 ): 1606 – 1613 . OpenUrl CrossRef PubMed 50. Maclean , Andrew E. , et al. “ Absence of complex I is associated with diminished respiratory chain function in European mistletoe .” Current Biology 28 . 10 ( 2018 ): 1614 – 1619 . OpenUrl CrossRef PubMed 51. Leger , M. , Kolisko , M. , Kamikawa , R. et al. Organelles that illuminate the origins of Trichomonas hydrogenosomes and Giardia mitosomes . Nat Ecol Evol 1 , 0092 ( 2017 ). doi: 10.1038/s41559-017-0092 OpenUrl CrossRef PubMed 52. Xu , Jiahui , et al. “ Exploring the evolution of anaerobes within ciliate class Prostomatea by transcriptomics .” Molecular Phylogenetics and Evolution ( 2025 ): 108345 . 53. de Graaf , Rob M. , et al. “ The hydrogenosomes of Psalteriomonas lanterna .” BMC Evolutionary Biology 9 ( 2009 ): 1 – 17 . OpenUrl CrossRef PubMed 54. Eldeeb , Mazzen H. , Lizeth J. Camacho Lopez , and Flavia Fontanesi . “ Mitochondrial respiratory supercomplexes of the yeast Saccharomyces cerevisiae .” IUBMB life 76 . 8 ( 2024 ): 485 – 504 . OpenUrl CrossRef PubMed 55. Müller , Miklós , et al. “ Biochemistry and evolution of anaerobic energy metabolism in eukaryotes .” Microbiology and Molecular Biology Reviews 76 . 2 ( 2012 ): 444 – 495 . OpenUrl Abstract / FREE Full Text 56. Tasbihi , Minoo , et al. “ Comparative mitochondrial proteomics of Leishmania tropica clinical isolates resistant and sensitive to meglumine antimoniate .” Parasitology Research 119 ( 2020 ): 1857 – 1871 . OpenUrl CrossRef PubMed 57. Schröder , Lucie , et al. “ The gene space of European mistletoe (Viscum album) .” The Plant Journal 109 . 1 ( 2022 ): 278 – 294 . OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted May 04, 2025. Download PDF Data/Code Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Convergent pathways of reductive mitochondrial evolution characterised with hypercubic inference Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Convergent pathways of reductive mitochondrial evolution characterised with hypercubic inference Robert C. Glastad , Iain G. Johnston bioRxiv 2025.04.29.651065; doi: https://doi.org/10.1101/2025.04.29.651065 Share This Article: Copy Citation Tools Convergent pathways of reductive mitochondrial evolution characterised with hypercubic inference Robert C. Glastad , Iain G. Johnston bioRxiv 2025.04.29.651065; doi: https://doi.org/10.1101/2025.04.29.651065 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Evolutionary Biology Subject Areas All Articles Animal Behavior and Cognition (7616) Biochemistry (17625) Bioengineering (13852) Bioinformatics (41825) Biophysics (21397) Cancer Biology (18524) Cell Biology (25417) Clinical Trials (138) Developmental Biology (13350) Ecology (19858) Epidemiology (2067) Evolutionary Biology (24277) Genetics (15581) Genomics (22459) Immunology (17698) Microbiology (40278) Molecular Biology (17134) Neuroscience (88400) Paleontology (666) Pathology (2823) Pharmacology and Toxicology (4812) Physiology (7632) Plant Biology (15106) Scientific Communication and Education (2042) Synthetic Biology (4281) Systems Biology (9807) Zoology (2266)