The Close Mutual Errors Regarding Node Dating

Many molecular dating studies rely on a few, sometimes poorly understood fossils equally historic catamenia priors to constrain nodes heights (ages) inward an ultrametric tree. But do the authors (peers, editors, in addition to – ultimately – readers) know what they do/has been done? Maybe, mayhap not; inward whatsoever illustration reading the papers tin move confusing. In this post, I'll attempt to hand a quick step-in.

The real regulation of node dating using fossils

It has been pointed out occassionally that many molecular clocks are also young. This critique is ofttimes truthful but overlooks what node dating does. The oldest (known) representative of a lineage (e.g. genus or family) is used to constrain the possible minimum stem (root) historic catamenia of that lineage. Since nosotros (typically) do non know how closed this oldest (and recognisable) representative is to the actual start fellow member of the lineage, i.e. the mutual ancestor (CA) of the lineage (the lineage's root), it tin exclusively inform a lower boundary (Fig. 1). Thus, node-dating-based estimates tin move expected to move underestimating inward most of the cases in addition to should ever move regarded in addition to treated equally minima. The closer the fossil(s) used equally constraint is (are) from the actual lineage-CA (root nodes), the less underestimating volition move the minimum historic catamenia estimates.

Fig. 1 H5N1 truthful (ultrametric) tree of a genus alongside 3 extant taxa. The genus' crown historic catamenia is defined past times the most recent mutual ancestor of the modern species; the genus' stalk historic catamenia past times the betoken at which the genus diverged from its sis clade.

Using A’, the oldest known fossil of lineage A, equally minimum historic catamenia constraint for the MRCA of the modern species A, C, in addition to D, 1 may infer a real immature in addition to underestimated deviation historic catamenia for the sis species C in addition to D or the genus’ stalk age. But the estimates – e.g. C in addition to D diverged before 2.5 Ma, the genus diverged latest at xx Ma – are non wrong, the existent deviation ages (10 Ma, xl Ma) are greater than the estimates.
New palaeontological prove may demo that lineages H5N1 in addition to CD were already acquaint at nearly xx Ma, in addition to accordingly our node-dating minimum estimates volition move exclusively slightly underestimating. Since many dating studies rely on conservatively assigned (regarding taxonomy) deep node constraints (not rarely taken from before dating studies), it is non overly surprising that estimates towards the tips tin move highly underestimating.

Fig. 2 Same tree, but older fossils that are closer to the nodes they tin constrain, which automatically leads to less underestimating minimum deviation ages

Sensible node dating requires straight input/control past times palaeontologists

Fact is: exclusively palaeontological experts (may) receive got the overview to create upward one's remove heed which fossil is a goodness candidate for a node-dating constraint. Any (node) dating report stands in addition to falls alongside the character of the known fossil record. To acquire substantially amend dating estimates, 1 primarily needs a amend understood fossil record. Nothing else.
Relatively immature fossils (A’, B, S1 inward Figs 1 in addition to 2) musts upshot inward (severely) underestimated (“too young”) deviation historic catamenia estimates.
Wrongly assigned/misinterpreted fossils naturally may upshot inward fundamentally incorrect estimates (Fig. 3).

Fig. 3 Same tree, x-axis directly reflecting development of the genus inside its morphospace. Showing essentially a D-type morphology, fossil S1 (Fig. 2) is labelled equally D', in addition to erroneously used past times the dater to constrain a much also high minimum historic catamenia of the MRCA of C in addition to D (= stalk historic catamenia of D).

Examples of odd, meaningless or fifty-fifty methodologically flawed dating estimates – related to lack of proper historic catamenia constraints or role of poorly understood molecular information (problematic topologies) – tin move found throughout literature, inward low- to high-impact journals. The most intriguing illustration for me was a newspaper past times Larson-Johnson (2016) on charge per unit of measurement shifts inward Fagales, published terminal yr inward confidentially peer-reviewed journal (Sauquet et al. 2012; the methodology is fine, but the information in addition to tested scenarios are not). In illustration you lot are interested inward to a greater extent than details, encounter the Supplemental Files S2, S3 (included inward this archive) of Grímsson et al. (2016), in addition to our discussion.

Furthermore, the palaeontologist may receive got a goodness thought whether a fossil represents a stalk or crown taxon of a lineage. This tin aid to acquire to a greater extent than useful estimates, fifty-fifty if it way to curvature the rules of node dating. The ground lies inward the somewhat dissimilar Definition of stalk in addition to crown when using Hennig’s phylogenetic classification or cladistics equally used inward molecular dating (Fig. 4).

Fig. 4 Same tree, alongside ii to a greater extent than fossil taxa. A*, the start representative of the lineage leading to H5N1 (earliest precursor of A), in addition to S2, a crown fossil (according Definition of Doyle & Donoghue 1993), but business office of the stalk (root branch) inward a node-dating framework. Both fossils would hand sensible estimates when used to constrain the genus' crown historic catamenia (MRCA modern species), but inward illustration of S2 they would move overestimating. Note that inward the existent world, it may move impossible to create upward one's remove heed whether a fossil represents A* or S2.

Fossil S2 shows all (syn)apomorphies of the (modern) genus, thus, it tin move regarded equally a crown taxon. It would brand a sensible constraint for the historic catamenia (absolute, non minimum) of the most recent mutual ancestor (MRCA) of all modern taxa (A, C, D), i.e. render a straight constraint for the crown historic catamenia of the genus. But since it is business office of the stalk (cladistically speaking) in addition to cannot move assigned to 1 of the subclades within the genus (A, BCD, CD, C, D), it tin exclusively move used to inform a (much also young) minimum stalk historic catamenia for the genus itself next the measure protocol of node dating. Note that when nosotros – against the rules – role S2 to constrain the genus' crown historic catamenia all subsequent estimates may move (a bit) also old! However, inward reality, it is much to a greater extent than probable that such an early on crown taxon is non the real terminal mutual ancestor of all modern species or stalk taxon (S2), but either a precursor (A*) or early on (extinct) sibling (like B), hence, probable provides a best-as-possible minimum historic catamenia constraint for the (modern) genus’ crown age. H5N1 real-world illustration is Fagus langevinii [PDF], used past times Sauquet et al. (2012) – methodologically right – equally much also immature minimum stalk (root) historic catamenia constraint for the lineage leading to the beech trees, Fagus. Regarding its morphology in addition to the fossil tape inward general, the fossil is to a greater extent than probable to inform a sensible crown historic catamenia of the (modern) genus (Denk & Grimm 2009; Grímsson et al. 2016; Renner et al. 2016).

Lineage sampling in addition to topological ambiguity

Something ofttimes overlooked inward node dating studies, in addition to their interpretation, is lineage sampling in addition to the bias because of the used topology. Let’s assume that nosotros don’t receive got whatsoever information on C in addition to D inward our tree, in addition to exclusively A-type/-similar fossils (Fig. 5). There is no genus crown anymore, simply a unmarried taxon. Following the node dating rules, A* or S1 tin exclusively inform a minimum stalk (root) historic catamenia of the genus (represented past times A), which volition move also young.

Fig. v Same tree, but C in addition to D receive got non been sampled for the analysis. Accordingly A*, which unremarkably would render a best-possible constraint for the genus' crown age, exclusively provides a miserable constraint for the genus' stalk age.

As before long equally nosotros add together C or D to the dataset, some other species of the genus, A* does inform the stalk (root) historic catamenia of lineage A, i.e. 1 node up, in addition to tin render us alongside quite a sensible historic catamenia prior (fix point; Fig. 3). However, if our inferred tree resolves C equally sis of H5N1 but non D, we'll assign A* to an artificial, also immature node.

Fig. half dozen A partly incorrect tree (just shifted the genus' root); directly A* constrains the MRCA of H5N1 in addition to C, but non anymore of all modern species A, C, in addition to D equally before.

H5N1 few real-world examples illustrating these issues.
Liu et al. (2014) argued that an Eocene Alnus fossil is a fellow member of 1 of the subgenera (subgenus Alnus), in addition to concluded that the Alnus subgenera were already established past times that time, i.e. crown grouping radiations started, contrasting our dating estimates inferring a xv Ma younger crown historic catamenia (Grimm & Renner 2013). But inward fact, there’s no conflict, equally our information laid (harvested from cistron banks) exclusively included members of subgenus Alnus – equally it turned out later. Hence, our crown historic catamenia exclusively refers to the subgenus Alnus, but non the genus Alnus itself, in addition to stands unchallenged.
Not exclusively the Fagales charge per unit of measurement shifts estimated past times Larson-Johnson (2016) are based on a pretty incorrect topology, inward special when it comes to intra-family relationships (poor genetic information selection). Similar problems tin move found inward the all-Fagales studies of Xing et al. (2014) in addition to Xiang et al. (2014). Although all 3 studies role partly the same fossils to constrain node ages – fossils also included past times Sauquet et al. (2012) – they identify them at dissimilar nodes (see Fig. S2-1 inward the supplement to Grímsson et al., 2016). Hence, the same fossils are used to inform the historic catamenia of (substantially) dissimilar MRCAs, but all in-line alongside node dating procedure! Sauquet et al. in addition to Larson-Johnson eliminated some of the topological ambiguity imminent from their information sets past times leaving out a span of genera. Thus, assigning fossil-based historic catamenia constraints to, actually, also deep nodes inward their tree, similar to what I pull inward Fig. 5.

If you lot must do node dating…

The fossilized birth-death dating (FBD; Heath, Huelsenbeck & Stadler 2014) is clearly superior (in regulation in addition to practise) in addition to if you lot receive got genuinely goodness morphological data, total-evidence dating (Ronquist et al. 2012) may move an alternative (not for plants, I'm afraid). Nonetheless, classic node dating may move inevitable inward many cases. So, here’s some advice what to do (and what peers should inquire when judging a node-dating paper):

• Contact a palaeontologist who worked on the grouping from which you lot select your dating constraints in addition to has an overview most fossil record. (If you’re lucky, you lot acquire a total fossil tape that allows you lot to do FBD.)
• If nobody is available, many modern groups are however poorly studied regarding their fossil records in addition to palaeontological cognition is somewhat dying out (at to the lowest degree the number of palaeobotanists in addition to positions is steadily declining), make a proper literature search (i.e. checking master copy palaeontological literature, non other dating papers), in addition to put upward a tabular array listing the fossil records. Provide that tabular array equally supplement (open data), in addition to so others tin elaborate on it (always a prissy gesture for hereafter research).
• Given the uncertainty most the actual positions of fossils inward the phylogeny (is it a A*, B or S1?), 1 should play simply about alongside dissimilar constraints. Rather than using all fossils to constrain many nodes inward your chronogram at once, run several chronograms each using simply a unmarried constraint in addition to compare the results. If the dissimilar fossil constraints halt upward alongside similar estimates, they can’t move also wrong. This also serves equally essay out to encounter which fossil doesn’t lucifer at all (maybe because they receive got been misinterpreted, mayhap because the molecular-inferred topology has some issues). But if you lot role them upward inward 1 unmarried run, you lot receive got no command on their biases.
• Same for topologies. The era of Big Data, next-generation sequencing in addition to (beauti)fully resolved, unambiguous phylogenies may move dawning, but most electrical flow information sets used for dating are however oligogene data. They tin include (substantial) internal conflict (e.g. the information used inward the all-Fagales studies mentioned above) in addition to may move supporting to a greater extent than than 1 topological alternative; in addition to eventually resulting inward a Bayesian highest posterior probability (PP) tree that has some strange branching patterns (watch for depression PP, but also branches alongside high PP in addition to depression bootstrap support). Make sure enough you lot receive got a proper tree or several of them in addition to prepare the topologies for the dating analysis rather than move out it all to the BEAST.

One terminal banking concern annotation on the data: if you’re doing dating on genera (species-level), you lot naturally desire to include equally many species equally possible, but you demand goodness data, also (ideally no information gaps, information alongside sufficient diversity). If you’re doing families or higher hierarchies, whatsoever ii genera volition hand you lot a crown in addition to let estimating a crown age, but don’t confuse this alongside the actual identify unit of measurement crown (age); encounter e.g. some of the youngest alternatives for deviation ages provided on the otherwise bright Angiosperm Phylogeny Website (Stevens 2001 onwards). To acquire a (methodologically) meaningful identify unit of measurement crown historic catamenia 1 needs to sample all genera/lineages of a family, or at to the lowest degree the (putatively) earliest diverging branches.

And ever remember: Whatever the result, it’s all simply minimum estimates. But, when thoughfully done, they may move closed to the actual betoken of divergence.

Cited literature

Denk T, Grimm GW. 2009. The biogeographic history of beech trees. Review of Palaeobotany in addition to Palynology 158:83–100.
Grimm GW, Renner SS. 2013. Harvesting GenBank for a Betulaceae supermatrix, in addition to a novel chronogram for the family. Botanical Journal of the Linnéan Society 172:465–477.
Heath TA, Huelsenbeck JP, Stadler T. 2014. The fossilized birth–death physical care for for coherent calibration of divergence-time estimates. Proceedings of the National Academy of Sciences 111:E2957–E2966.
Liu X, Manchester SR, Jin J. 2014. Alnus subgenus Alnus in the Eocene of Western North America based on leaves, associated catkins, pollen, in addition to fruits. American Journal of Botany 101:1925–1943.
Renner SS, Grimm GW, Kapli P, Denk T. 2016. Species relationships in addition to deviation times inward beeches: New insights from the inclusion of 53 immature in addition to onetime fossils inward a birth-death clock model. Philosophical Transactions of the Royal Society B DOI:10.1098/rstb.2015.0135.
Ronquist F, Klopfstein S, Vilhelmsen L, Schulmeister S, Murray DL, Rasnitsyn AP. 2012. H5N1 total-evidence approach to dating alongside fossils, applied to the early on radiations of the hymenoptera. Systematic Biology 61:973–999.
Stevens PF. 2001 onwards. Angiosperm Phylogeny Website. Version 8, June 2007 [and to a greater extent than or less continuously updated since]. Available at http://www.mobot.org/MOBOT/research/APweb/ (accessed 30/03/2017.
Xiang X-G, Wang W, Li R-Q, Lin L, Liu Y, Zhou Z-K, Li Z-Y, Chen Z-D. 2014. Large-scale phylogenetic analyses disclose fagalean diversification promoted past times the interplay of diaspores in addition to environments inward the Paleogene. Perspectives inward Plant Ecology, Evolution in addition to Systematics 16:101–110.
Xing Y, Onstein RE, Carter RJ, Stadler T, Linder HP. 2014. Fossils in addition to large molecular phylogeny demo that the development of species richness, generic diversity, in addition to turnover rates are disconnected. Evolution 68:2821–2832.
Grímsson F, Grimm GW, Zetter R, Denk T. 2016. Cretaceous in addition to Paleogene Fagaceae from North America in addition to Greenland: prove for a Late Cretaceous split upward betwixt Fagus and the remaining Fagaceae. Acta Palaeobotanica 56:247–305. http://dx.doi.org/10.1515/acpa-2016-0016
Larson-Johnson K. 2016. Phylogenetic investigation of the complex evolutionary history of dispersal fashion in addition to diversification rates across living in addition to fossil Fagales. New Phytologist 209:418–435.
Sauquet H, Ho SY, Gandolfo MA, Hashemite Kingdom of Jordan GJ, Wilf P, Cantrill DJ, Bayly MJ, Bromham L, Brown GK, Carpenter RJ, Lee DM, Murphy DJ, Sniderman JM, Udovicic F. 2012. Testing the affect of calibration on molecular deviation times using a fossil-rich group: the illustration of Nothofagus (Fagales). Systematic Biology 61:289–313.