Viral spillover risk increases with climate change in High Arctic lake sediments.

The host spectrum of viruses is quite diverse, as they can sustainedly infect a few species to several phyla. When confronted with a new host, a virus may even infect it and transmit sustainably in this new host, a process called 'viral spillover'. However, the risk of such events is difficult to quantify. As climate change is rapidly transforming environments, it is becoming critical to quantify the potential for spillovers. To address this issue, we resorted to a metagenomics approach and focused on two environments, soil and lake sediments from Lake Hazen, the largest High Arctic freshwater lake in the world. We used DNA and RNA sequencing to reconstruct the lake's virosphere in both its sediments and soils, as well as its range of eukaryotic hosts. We then estimated the spillover risk by measuring the congruence between the viral and the eukaryotic host phylogenetic trees, and show that spillover risk increases with runoff from glacier melt, a proxy for climate change. Should climate change also shift species range of potential viral vectors and reservoirs northwards, the High Arctic could become fertile ground for emerging pandemics.

Direct Evidence of an Increasing Mutational Load in Humans.

The extent to which selection has shaped present-day human populations has attracted intense scrutiny, and examples of local adaptations abound. However, the evolutionary trajectory of alleles that, today, are deleterious has received much less attention. To address this question, the genomes of 2,062 individuals, including 1,179 ancient humans, were reanalyzed to assess how frequencies of risk alleles and their homozygosity changed through space and time in Europe over the past 45,000 years. Although the overall deleterious homozygosity has consistently decreased, risk alleles have steadily increased in frequency over that period of time. Those that increased most are associated with diseases such as asthma, Crohn disease, diabetes, and obesity, which are highly prevalent in present-day populations. These findings may not run against the existence of local adaptations but highlight the limitations imposed by drift and population dynamics on the strength of selection in purging deleterious mutations from human populations.

Identifying genetic determinants of complex phenotypes from whole genome sequence data.

BACKGROUND: A critical goal in biology is to relate the phenotype to the genotype, that is, to find the genetic determinants of various traits. However, while simple monofactorial determinants are relatively easy to identify, the underpinnings of complex phenotypes are harder to predict. While traditional approaches rely on genome-wide association studies based on Single Nucleotide Polymorphism data, the ability of machine learning algorithms to find these determinants in whole proteome data is still not well known. RESULTS: To better understand the applicability of machine learning in this case, we implemented two such algorithms, adaptive boosting (AB) and repeated random forest (RRF), and developed a chunking layer that facilitates the analysis of whole proteome data. We first assessed the performance of these algorithms and tuned them on an influenza data set, for which the determinants of three complex phenotypes (infectivity, transmissibility, and pathogenicity) are known based on experimental evidence. This allowed us to show that chunking improves runtimes by an order of magnitude. Based on simulations, we showed that chunking also increases sensitivity of the predictions, reaching 100% with as few as 20 sequences in a small proteome as in the influenza case (5k sites), but may require at least 30 sequences to reach 90% on larger alignments (500k sites). While RRF has less specificity than random forest, it was never <50%, and RRF sensitivity was significantly higher at smaller chunk sizes. We then used these algorithms to predict the determinants of three types of drug resistance (to Ciprofloxacin, Ceftazidime, and Gentamicin) in a bacterium, Pseudomonas aeruginosa. While both algorithms performed well in the case of the influenza data, results were more nuanced in the bacterial case, with RRF making more sensible predictions, with smaller errors rates, than AB. CONCLUSIONS: Altogether, we demonstrated that ML algorithms can be used to identify genetic determinants in small proteomes (viruses), even when trained on small numbers of individuals. We further showed that our RRF algorithm may deserve more scrutiny, which should be facilitated by the decreasing costs of both sequencing and phenotyping of large cohorts of individuals.

How the Central American Seaway and an Ancient Northern Passage Affected Flatfish Diversification.

While the natural history of flatfish has been debated for decades, the mode of diversification of this biologically and economically important group has never been elucidated. To address this question, we assembled the largest molecular data set to date, covering > 300 species (out of ca. 800 extant), from 13 of the 14 known families over nine genes, and employed relaxed molecular clocks to uncover their patterns of diversification. As the fossil record of flatfish is contentious, we used sister species distributed on both sides of the American continent to calibrate clock models based on the closure of the Central American Seaway (CAS), and on their current species range. We show that flatfish diversified in two bouts, as species that are today distributed around the equator diverged during the closure of CAS, whereas those with a northern range diverged after this, hereby suggesting the existence of a postCAS closure dispersal for these northern species, most likely along a trans-Arctic northern route, a hypothesis fully compatible with paleogeographic reconstructions.

Flight costs in volant vertebrates: A phylogenetically-controlled meta-analysis of birds and bats.

Flight costs play an important role in determining the behavior, ecology, and physiology of birds and bats. Mechanical flight costs can be estimated from aerodynamics. However, measured metabolic flight costs (oxygen consumption rate) are less accurately predicted by flight theory, either because of (1) variation in flight efficiency across species, (2) variation in how basal costs interact with flight costs or (3) methodological biases. To tease apart these three hypotheses, we conducted a phylogenetically-controlled meta-analysis based on data from birds and bats. Birds doing short flights in a lab had higher metabolic rates than those with sustained flapping flight. In turn, species that used sustained flapping flight had a higher metabolic rate than those that flew primarily via gliding. Models accounting for relatedness (phylogeny) explained the data better than those that did not, which is congruent with the idea that several different flight Bauplans have evolved within birds and bats. Focusing on species with sustained flapping flight, for which more data are currently available, we found that flight cost estimates were not affected by measurement methods in both birds and bats. However, efficiency increased with body mass and decreased with flight speed in both birds and bats. Basal metabolic rate was additive to flight metabolic rate in bats but not birds. We use these results to derive an equation for estimating metabolic flight costs of birds and bats that includes variation in whole animal efficiency with flight speed and body mass.

Both Epistasis and Diversifying Selection Drive the Structural Evolution of the Ebola Virus Glycoprotein Mucin-Like Domain.

UNLABELLED: Throughout the last 3 decades, Ebola virus (EBOV) outbreaks have been confined to isolated areas within Central Africa; however, the 2014 variant reached unprecedented transmission and mortality rates. While the outbreak was still under way, it was reported that the variant leading up to this outbreak evolved faster than previous EBOV variants, but evidence for diversifying selection was undetermined. Here, we test this selection hypothesis and show that while previous EBOV outbreaks were preceded by bursts of diversification, evidence for site-specific diversifying selection during the emergence of the 2014 EBOV clade is weak. However, we show strong evidence supporting an interplay between selection and correlated evolution (epistasis), particularly in the mucin-like domain (MLD) of the EBOV glycoprotein. By reconstructing ancestral structures of the MLD, we further propose a structural mechanism explaining how the substitutions that accumulated between 1918 and 1969 distorted the MLD, while more recent epistatic substitutions restored part of the structure, with the most recent substitution being adaptive. We suggest that it is this complex interplay between weak selection, epistasis, and structural constraints that has shaped the evolution of the 2014 EBOV variant. IMPORTANCE: The role that selection plays in the emergence of viral epidemics remains debated, particularly in the context of the 2014 EBOV outbreak. Most critically, should such evidence exist, it is generally unclear how this relates to function and increased virulence. Here, we show that the viral lineage leading up to the 2014 outbreak underwent a complex interplay between selection and correlated evolution (epistasis) in a protein region that is critical for immune evasion. We then reconstructed the three-dimensional structure of this domain and showed that the initial mutations in this lineage deformed the structure, while subsequent mutations restored part of the structure. Along this mutational path, the first and last mutations were adaptive, while the intervening ones were epistatic. Altogether, we provide a mechanistic model that explains how selection and epistasis acted on the structural constraints that materialized during the 2014 EBOV outbreak.

Automatic detection of rate change in large data sets with an unsupervised approach: the case of influenza viruses.

Influenza viruses evolve at such a high rate that vaccine recommendations need to be changed, but not quite on a regular basis. This observation suggests that the rate of evolution of these viruses is not constant through time, which begs the question as to when such rate changes occur, if they do so independently of the host in which they circulate and (or) independently of their subtype. To address these outstanding questions, we introduce a novel heuristics, Mclust*, that is based on a two-tier clustering approach in a phylogenetic context to estimate (i) absolute rates of evolution and (ii) when rate change occurs. We employ the novel approach to compare the two influenza surface proteins, hemagglutinin and neuraminidase, that circulated in avian, human, and swine hosts between 1960 and 2014 in two subtypes: H3N2 and H1N1. We show that the algorithm performs well in most conditions, accounting for phylogenetic uncertainty by means of bootstrapping and scales up to analyze very large data sets. Our results show that our approach is robust to the time-dependent artifact of rate estimation, and confirm pervasive punctuated evolution across hosts and subtypes. As such, the novel approach can potentially detect when vaccine composition needs to be updated.

Comparative dynamics and distribution of influenza drug resistance acquisition to protein m2 and neuraminidase inhibitors.

Although efficient influenza vaccines are designed on a regular basis, the only protection of human populations against an unforeseen virus such as during the H1N1 pandemic in 2009 might be antiviral drugs. Adamantanes and neuraminidase inhibitors (Oseltamivir) represent two classes of such drugs that target the viral matrix protein 2 and neuraminidase, respectively. Although the emergence of resistance to both drugs has been described, the timing and spread of the acquisition of either single or dual resistances by different hosts is still unclear. Using a multilayered phylogenetic approach based on relaxed molecular clocks and large-scale maximum likelihood approaches, we show that Adamantane resistance evolved multiple times in various subtypes and hosts, possibly in breeding contexts (swine); and Oseltamivir resistance was also found in different subtypes and hosts, but its transmission is only sustained in humans. Furthermore, the dynamics of the emergence of antiviral resistance were examined for each drug. This showed that although the first mutations conferring resistance to Adamantanes precede US Food and Drug Administration (FDA) approval, general resistance emerged 15-38 years post-drug approval. This is in contrast to Oseltamivir resistance mutations that emerged at most 7 years after FDA approval of the drug. This study demonstrates the power of large-scale analyses to uncover and monitor the emergence dynamics of drug resistance.

Genome analyses suggest the presence of polyploidy and recent human-driven expansions in eight global populations of the honeybee pathogen Nosema ceranae.

Nosema ceranae is a microsporidian pathogen whose infections have been associated with recent global declines in the populations of western honeybees (Apis mellifera). Despite the outstanding economic and ecological threat that N. ceranae may represent for honeybees worldwide, many aspects of its biology, including its mode of reproduction, propagation and ploidy, are either very unclear or unknown. In the present study, we set to gain knowledge in these biological aspects by re-sequencing the genome of eight isolates (i.e. a population of spores isolated from one single beehive) of this species harvested from eight geographically distant beehives, and by investigating their level of polymorphism. Consistent with previous analyses performed using single gene sequences, our analyses uncovered the presence of very high genetic diversity within each isolate, but also very little hive-specific polymorphism. Surprisingly, the nature, location and distribution of this genetic variation suggest that beehives around the globe are infected by a population of N. ceranae cells that may be polyploid (4n or more), and possibly clonal. Lastly, phylogenetic analyses based on genome-wide single-nucleotide polymorphism data extracted from these parasites and mitochondrial sequences from their hosts all failed to support the current geographical structure of our isolates.

Dual foraging and pair coordination during chick provisioning by Manx shearwaters: empirical evidence supported by a simple model.

The optimal allocation of time and energy between one's own survival and offspring survival is critical for iteroparous animals, but creates a conflict between what maximises the parent's fitness and what maximises fitness of the offspring. For central-place foragers, provisioning strategies may reflect this allocation, while the distance between central-places and foraging areas may influence the decision. Nevertheless, few studies have explored the link between life history and foraging in the context of resource allocation. Studying foraging behaviour alongside food load rates to chicks provides a useful system for understanding the foraging decisions made during parent-offspring conflict. Using simultaneously deployed GPS and time-depth recorders, we examined the provisioning strategies in free-living Manx shearwaters Puffinus puffinus, which were caring for young. Our results showed a bimodal pattern, where birds alternate short and long trips. Short trips were associated with higher feeding frequency and larger meals than long trips, suggesting that long trips were performed for self-feeding. Furthermore, most foraging was carried out within 100 km of sea fronts. A simple model based on patch quality and travel time shows that for Manx shearwaters combining chick feeding and self-maintenance, bimodal foraging trip durations optimise feeding rates.

Predicting the emergence of H3N2 influenza viruses reveals contrasted modes of evolution of HA and NA antigens.

Vaccine design for rapidly changing viruses is based on empirical surveillance of strains circulating in a given season to assess those that will most likely spread during the next season. The choice of which strains to include in the vaccine is critical, as an erroneous decision can lead to a nonimmunized human population that will then be at risk in the face of an epidemic or, worse, a pandemic. Here, we present the first steps toward a very general phylogenetic approach to predict the emergence of novel viruses. Our genomic model builds upon natural features of viral evolution such as selection and recombination / reassortment, and incorporates episodic bursts of evolution and or of recombination. As a proof-of-concept, we assess the performance of this model in a retrospective study, focusing: (i) on the emergence of an unexpected H3N2 influenza strain in 2007, and (ii) on a longitudinal design. Based on the analysis of hemagglutinin (HA) and neuraminidase (NA) genes, our results show a lack of predictive power in both experimental designs, but shed light on the mode of evolution of these two antigens: (i) supporting the lack of significance of recombination in the evolution of this influenza virus, and (ii) showing that HA evolves episodically while NA changes gradually.

Habitat-specific type I polyketide synthases in soils and street sediments.

Actinomycetes produce many pharmaceutically useful compounds through type I polyketide biosynthetic pathways. Soil has traditionally been an important source for these actinomycete-derived pharmaceuticals. As the rate of antibiotic discovery has decreased and the incidence of antibiotic resistance has increased, researchers have looked for alternatives to soil for bioprospecting. Street sediment, where actinomycetes make up a larger fraction of the bacterial population than in soil, is one such alternative environment. To determine if these differences in actinomycetal community structure are reflected in type I polyketide synthases (PKSI) distribution, environmental DNA from soils and street sediments was characterized by sequencing amplicons of PKSI-specific PCR primers. Amplicons covered two domains: the last 80 amino acids of the ketosynthase (KS) domain and the first 240 amino acids of the acyltransferase (AT) domain. One hundred and ninety clones from ten contrasting soils from six regions and nine street sediments from six cities were sequenced. Twenty-five clones from two earthworm-affected samples were also sequenced. UniFrac lineage-specific analysis identified two clades that clustered with actinomycetal GenBank matches that were street sediment-specific, one similar to the PKSI segment of the mycobactin siderophore involved in mycobacterial virulence. A clade of soil-specific sequences clustered with GenBank matches from the ambruticin and jerangolid pathways of Sorangium cellulosum. All three of these clades were found in sites >700 km apart. Street sediments are enriched in actinomycetal PKSIs. Non-actinomycetal PKSI pathways may be more chemically diverse than actinomycetal PKSIs. Common soil and street sediment PKIs are globally distributed.

Genetic diversity as a marker for timing infection in HIV-infected patients: evaluation of a 6-month window and comparison with BED.

BACKGROUND: It has been reported that the increase in human immunodeficiency virus (HIV) sequence diversity in drug resistance surveillance specimens may be used to classify the duration of HIV infection as <1 or >1 year. We describe a mixed base classifier (MBC) optimized to categorize the duration of subtype B infections as <6 or >6 months on the basis of sequences for drug resistance surveillance specimens and compared MBC findings with those of serologic methods. METHODS: The behavior of the MBC was examined across a range of thresholds for calling mixed bases. MBC performance was then evaluated using either complete pol sequences or sites reflecting evolutionary pressures (HLA selection sites, sites that increased in entropy over the course of infection, and codon positions). RESULTS: The MBC performance was optimal when secondary peaks on the sequencing chromatogram accounted for at least 15% of the area of primary peaks. A cutoff of <0.45% mixed bases in the pol region best identified recent infections (sensitivity = 82.7%, specificity = 78.8%), with improvement achieved by analyzing only sites that increased in entropy. CONCLUSIONS: In an extended data set of 1354 specimens classified by BED, the optimized MBC performed significantly better than a simple MBC (agreement, 68.98% vs 67.13%). If further validated, the MBC may prove beneficial for detecting recent infection and estimating the incidence of HIV infection.

Machine Learning Algorithms Associate Case Numbers with SARS-CoV-2 Variants Rather Than with Impactful Mutations.

During the SARS-CoV-2 pandemic, much effort has been geared towards creating models to predict case numbers. These models typically rely on epidemiological data, and as such overlook viral genomic information, which could be assumed to improve predictions, as different variants show varying levels of virulence. To test this hypothesis, we implemented simple models to predict future case numbers based on the genomic sequences of the Alpha and Delta variants, which were co-circulating in Texas and Minnesota early during the pandemic. Sequences were encoded, matched with case numbers at a future time based on collection date, and used to train two algorithms: one based on random forests and one based on a feed-forward neural network. While prediction accuracies were ≥93%, explainability analyses showed that the models were not associating case numbers with mutations known to have an impact on virulence, but with individual variants. This work highlights the necessity of gaining a better understanding of the data used for training and of conducting explainability analysis to assess whether model predictions are misleading.

Mining-impacted rice paddies select for Archaeal methylators and reveal a putative (Archaeal) regulator of mercury methylation.

Methylmercury (MeHg) is a microbially produced neurotoxin derived from inorganic mercury (Hg), which accumulation in rice represents a major health concern to humans. However, the microbial control of MeHg dynamics in the environment remains elusive. Here, leveraging three rice paddy fields with distinct concentrations of Hg (Total Hg (THg): 0.21-513 mg kg-1 dry wt. soil; MeHg: 1.21-6.82 ng g-1 dry wt. soil), we resorted to metagenomics to determine the microbial determinants involved in MeHg production under contrasted contamination settings. We show that Hg methylating Archaea, along with methane-cycling genes, were enriched in severely contaminated paddy soils. Metagenome-resolved Genomes of novel putative Hg methylators belonging to Nitrospinota (UBA7883), with poorly resolved taxonomy despite high completeness, showed evidence of facultative anaerobic metabolism and adaptations to fluctuating redox potential. Furthermore, we found evidence of environmental filtering effects that influenced the phylogenies of not only hgcA genes under different THg concentrations, but also of two housekeeping genes, rpoB and glnA, highlighting the need for further experimental validation of whether THg drives the evolution of hgcAB. Finally, assessment of the genomic environment surrounding hgcAB suggests that this gene pair may be regulated by an archaeal toxin-antitoxin (TA) system, instead of the more frequently found arsR-like genes in bacterial methylators. This suggests the presence of distinct hgcAB regulation systems in bacteria and archaea. Our results support the emerging role of Archaea in MeHg cycling under mining-impacted environments and shed light on the differential control of the expression of genes involved in MeHg formation between Archaea and Bacteria.

Long-Term Tillage and Crop Rotation Regimes Reshape Soil-Borne Oomycete Communities in Soybean, Corn, and Wheat Production Systems.

Soil-borne oomycetes include devastating plant pathogens that cause substantial losses in the agricultural sector. To better manage this important group of pathogens, it is critical to understand how they respond to common agricultural practices, such as tillage and crop rotation. Here, a long-term field experiment was established using a split-plot design with tillage as the main plot factor (conventional tillage (CT) vs. no till (NT), two levels) and rotation as the subplot factor (monocultures of soybean, corn, or wheat, and corn-soybean-wheat rotation, four levels). Post-harvest soil oomycete communities were characterized over three consecutive years (2016-2018) by metabarcoding the Internal Transcribed Spacer 1 (ITS1) region. The community contained 292 amplicon sequence variants (ASVs) and was dominated by Globisporangium spp. (85.1% in abundance, 203 ASV) and Pythium spp. (10.4%, 51 ASV). NT decreased diversity and community compositional structure heterogeneity, while crop rotation only affected the community structure under CT. The interaction effects of tillage and rotation on most oomycetes species accentuated the complexity of managing these pathogens. Soil and crop health represented by soybean seedling vitality was lowest in soils under CT cultivating soybean or corn, while the grain yield of the three crops responded differently to tillage and crop rotation regimes.

Extreme diversity in noncalcifying haptophytes explains a major pigment paradox in open oceans.

The current paradigm holds that cyanobacteria, which evolved oxygenic photosynthesis more than 2 billion years ago, are still the major light harvesters driving primary productivity in open oceans. Here we show that tiny unicellular eukaryotes belonging to the photosynthetic lineage of the Haptophyta are dramatically diverse and ecologically dominant in the planktonic photic realm. The use of Haptophyta-specific primers and PCR conditions adapted for GC-rich genomes circumvented biases inherent in classical genetic approaches to exploring environmental eukaryotic biodiversity and led to the discovery of hundreds of unique haptophyte taxa in 5 clone libraries from subpolar and subtropical oceanic waters. Phylogenetic analyses suggest that this diversity emerged in Paleozoic oceans, thrived and diversified in the permanently oxygenated Mesozoic Panthalassa, and currently comprises thousands of ribotypic species, belonging primarily to low-abundance and ancient lineages of the "rare biosphere." This extreme biodiversity coincides with the pervasive presence in the photic zone of the world ocean of 19'-hexanoyloxyfucoxanthin (19-Hex), an accessory photosynthetic pigment found exclusively in chloroplasts of haptophyte origin. Our new estimates of depth-integrated relative abundance of 19-Hex indicate that haptophytes dominate the chlorophyll a-normalized phytoplankton standing stock in modern oceans. Their ecologic and evolutionary success, arguably based on mixotrophy, may have significantly impacted the oceanic carbon pump. These results add to the growing evidence that the evolution of complex microbial eukaryotic cells is a critical force in the functioning of the biosphere.

Testing alternative hypotheses on the origin and speciation of Hawaiian katydids.

BACKGROUND: Hawaiian Islands offer a unique and dynamic evolutionary theatre for studying origin and speciation as the islands themselves sequentially formed by erupting undersea volcanos, which would subsequently become dormant and extinct. Such dynamics have not been used to resolve the controversy surrounding the origin and speciation of Hawaiian katydids in the genus Banza, whose ancestor could be from either the Old-World genera Ruspolia and Euconocephalus, or the New World Neoconocephalus. To address this question, we performed a chronophylogeographic analysis of Banza species together with close relatives from the Old and New Worlds. RESULTS: Based on extensive dated phylogeographic analyses of two mitochondrial genes (COX1 and CYTB), we show that our data are consistent with the interpretation that extant Banza species resulted from two colonization events, both by katydids from the Old World rather than from the New World. The first event was by an ancestral lineage of Euconocephalus about 6 million years ago (mya) after the formation of Nihoa about 7.3 mya, giving rise to B. nihoa. The second colonization event was by a sister lineage of Ruspolia dubia. The dating result suggests that this ancestral lineage first colonized an older island in the Hawaiian-Emperor seamount chain before the emergence of Hawaii Islands, but colonized Kauai after its emergence in 5.8 mya. This second colonization gave rise to the rest of the Banza species in two major lineages, one on the older northwestern islands, and the other on the newer southwestern islands. CONCLUSION: Chronophylogeographic analyses with well-sampled taxa proved crucial for resolving phylogeographic controversies on the origin and evolution of species colonizing a new environment.

The nonadaptive nature of the H1N1 2009 Swine Flu pandemic contrasts with the adaptive facilitation of transmission to a new host.

BACKGROUND: The emergence of the 2009 H1N1 Influenza pandemic followed a multiple reassortment event from viruses originally circulating in swines and humans, but the adaptive nature of this emergence is poorly understood. RESULTS: Here we base our analysis on 1180 complete genomes of H1N1 viruses sampled in North America between 2000 and 2010 in swine and human hosts. We show that while transmission to a human host might require an adaptive phase in the HA and NA antigens, the emergence of the 2009 pandemic was essentially nonadaptive. A more detailed analysis of the NA protein shows that the 2009 pandemic sequence is characterized by novel epitopes and by a particular substitution in loop 150, which is responsible for a nonadaptive structural change tightly associated with the emergence of the pandemic. CONCLUSIONS: Because this substitution was not present in the 1918 H1N1 pandemic virus, we posit that the emergence of pandemics is due to epistatic interactions between sites distributed over different segments. Altogether, our results are consistent with population dynamics models that highlight the epistatic and nonadaptive rise of novel epitopes in viral populations, followed by their demise when the resulting virus is too virulent.

A time line of the environmental genetics of the haptophytes.

The use of genomic data and the rise of phylogenomics have radically changed our view of the eukaryotic tree of life at a high taxonomic level by identifying 4-6 "supergroups." Yet, our understanding of the evolution of key innovations within each of these supergroups is limited because of poor species sampling relative to the massive diversity encompassed by each supergroup. Here we apply a multigene approach that incorporates a wide taxonomic diversity to infer the time line of the emergence of strategic evolutionary transitions in the haptophytes, a group of ecologically and biogeochemically significant marine protists that belong to the Chromalveolata supergroup. Four genes (SSU, LSU, tufA, and rbcL) were extensively analyzed under several Bayesian models to assess the robustness of the phylogeny, particularly with respect to 1) data partitioning; 2) the origin of the genes (host vs. endosymbiont); 3) across-site rate variation; and 4) across-lineage rate variation. We show with a relaxed clock analysis that the origin of haptophytes dates back to 824 million years ago (Ma) (95% highest probability density 1,031-637 Ma). Our dating results show that the ability to calcify evolved earlier than previously thought, between 329 and 291 Ma, in the Carboniferous period and that the transition from mixotrophy to autotrophy occurred during the same time period. Although these two transitions precede a habitat change of major diversities from coastal/neritic waters to the pelagic realm (291-243 Ma, around the Permian/Triassic boundary event), the emergence of calcification, full autotrophy, and oceanic lifestyle seem mutually independent.