Rethinking phylogenetic comparative methods

In the paper

“Tests of adaptive evolution come in two flavors: those with an a priori hypothesis (or hypotheses) regarding which lineages belong to which distinct regimes based on ancestral state reconstruction of explanatory factors (Butler and King 2004;Beaulieu et al. 2012) and those where the locations of regime changes are themselves estimated along with the parameters of the OU process (Ingram and Mahler”

What the sources say

• ?Not checked — no full text of the sourcecites Butler and King 2004 source
• Supported

  The Forward Phase section says SURFACE places regime shifts using continuous trait data rather than a priori hypotheses. The Backward Phase section says its final model identifies where shifts occur and estimates the OU parameters, supporting the clause attributed to Ingram and Mahler. (LLM, conf=0.93)

  cites Ingram and Mahler 2013 source

In the paper

“For example, we highlighted in the introduction that we think HSMs (following Beaulieu et al. 2013;Beaulieu and O'Meara 2016) are a potentially powerful, and widely applicable solution, even though we did not consider these in detail here.”

What the sources say

• ?Not checked — no full text of the sourcecites Beaulieu et al. 2013 source
• Supported

  The discussion passage beginning “Our HiSSE framework spans this range” (page unavailable in our copy) says the method can detect diversification driven by either a focal trait or an unexamined factor, including cases where both matter. That supports the citing authors’ hedged description of hidden-state models as a potentially powerful, widely applicable solution. (LLM, conf=0.88)

  cites Beaulieu and O'Meara 2016 source

In the paper

“These tools have been embraced by fields such as computer science, epidemiology and the social sciences, but largely ignored by comparative biologists [rare exceptions are the recent introductions of phylogenetic path analysis by Hardenberg and Gonzalez-Voyer (2013) and the application of causal models to make inferences from paleontological time-series by Reitan and Liow (2017)].”

What the sources say

• Supported

  Split into 2 atoms — 2 confirmed, 0 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  cites Hardenberg and Gonzalez-Voyer (2013) source
• ?Not checked — no full text of the sourcecites Reitan and Liow (2017) source

In the paper

“This fundamental problem has sown doubts about the suitability and reliability of many methods in comparative biology (e.g., Losos 2011), even if it was not obvious that these issues were connected.”

What the sources say

• Supported

  In “Many Traits in Many Clades Do Not Exhibit Phylogenetic Effect,” Losos says incorporating phylogenetic information may be unnecessary or inappropriate for some traits and questions how errors from comparative methods compare with errors from ignoring phylogeny. The Conclusions add that phylogenetic methods that ignore variation in evolutionary processes often produce inaccurate or misleading results. These passages support the cited doubt about the suitability and reliability of comparative methods. (LLM, conf=0.88)

  cites Losos 2011 source

In the paper

“Furthermore, we also think that rigorous examination of goodness-of-fit and model adequacy following any comparative analysis is critical for finding unforeseen singular events driving signal in the data set (Garland et al. 1992;Boettiger et al. 2012;Slater and Pennell 2013;Pennell et al. 2015).”

What the sources say

• ?Not checked — no full text of the sourcecites Garland et al. 1992 source
• Partly supported

  The Introduction (§1.1; page unavailable) says comparative results can be difficult to interpret without assessing model fit, uncertainty, or power, supporting the value of rigorous evaluation. The cited paper does not say that post-analysis goodness-of-fit or model-adequacy checks reveal unforeseen singular events driving the signal in a dataset. (LLM, conf=0.82)

  cites Boettiger et al. 2012 sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites Slater and Pennell 2013 source

In the paper

“We also note that estimating a transformation for the errors (Pagel 1999;Freckleton et al. 2002) will not rescue the analysis; the estimated value of will lie between 0 and 1 and we have found both these more extreme cases (OLS and IC, respectively) to be susceptible.”

What the sources say

• ?Not checked — no full text of the sourcecites Pagel 1999 source
• ?Not checked — no full text of the sourcecites Freckleton et al. 2002 source

In the paper

“In the original Scales data set, there are weakly supported shifts in the clades leading to the sister group of Phrynosoma lizards, and the branch leading to Acanthodactylus scutellatus and Aspidoscelis tigris.”

In the paper

“Doing so recovers strongest support for the Phrynosoma-only model, intermediate support for the PE hypothesis, and very little weight on the reversible-jump hypothesis, which has no strongly supported shifts (Fig. 5).”

In the paper

“Here, we revisit Felsenstein's worst-case scenario in order to demonstrate that IC and Phylogenetic Generalized Least Squares (PGLS) do not completely address the problem that we tend to think they do-these methods are still susceptible to singular evolutionary events.”

In the paper

“In B and C, the correct model receives highest support with neither of the alternatives being well-supported.”

In the paper

“The shading of the points indicates cases where this assumption is violated, as outlying points with max(Q) values much greater than 1/T (where only 1 shift is expected) are much more likely to fall off the predicted line.”

In the paper

“In the original Scales data set, the Phrynosoma-only hypothesis receives the most weight (indicating a singular shift best explains the patterns observed in the data), while an intermediate weight is given to the PE hypothesis (which explains a good amount of the remaining variation).”

In the paper

“Therefore, both networks can produce data sets with phylogenetic signal in both body size and migratory behavior, and both can produce a data set with the distribution in Figure 9D (see Supplementary Material available on Dryad for additional details on generating Fig. 9).”

In the paper

“We estimate a high posterior probability on a regime shift in the genus Phrynosoma from the empirical data only (red circle), indicating that while the PE hypothesis explains some patterns in the data, it does not fully explain the shift present in the behaviorally and ecologically unique genus Phrynosoma.”

In the paper

“Rather, the great uncertainty in possible models is incorporated as a prior on the arrangement of shifts and is limited in explanatory power, something that researcher-driven biological hypotheses are much more capable of accomplishing.”

In the paper

“As discussed above, given a large enough tree such a priori hypotheses are likely to be strongly supported; if you can predict which one branch out of many will contain a shift then you may be on to something.”

In the paper

“As with Case Study I, these scenarios ultimately reduce to whether or not coincident unreplicated events are evidence of a causal link between traits (Maddison and FitzJohn 2015), a problem we will address in Case Study III.”

What the sources say

• ?Not checked — no full text of the sourcecites Maddison and FitzJohn 2015 source

In the paper

“Detecting a signal of evolutionary covariation is more difficult in discrete characters, but examining this situation isolates the recurring problem we have identified in the two previous Case Studies—whether or not coincident unreplicated events are evidence of causal links between traits.”

In the paper

“Therefore, the conclusions of Scales et al. (2009) hold up, even when accommodating evolutionary regime shifts unrelated to the factors being considered, although the weight of evidence appears to be larger than it actually is.”

What the sources say

• ?Not checked — no full text of the sourcecites Scales et al. (2009) source

In the paper

“Furthermore, models that actually describe such scenarios-like our”

In the paper

“Felsenstein also noted that the problem of nonindependence does not occur if”

What the sources say

• ?Not checked — no full text of the sourcecites Felsenstein 1985 source

In the paper

“In fact, discovering the location of wellsupported shifts on the phylogeny does not say anything about causation; it is merely a descriptive technique to find major features of the data where there is evidence that the parameters governing the dynamics of trait evolution have shifted on the phylogeny.”

In the paper

“Multioptima OU models have been widely used to test for the presence of shifts in evolutionary regimes (i.e., parts of the phylogeny with their own optima, or less commonly, their own strength of selection parameters).”

What the sources say

• ?Not checked — no full text of the sourcecites Butler and King 2004 source

In the paper

“We show here that allowing for unaccounted shifts can provide a stronger test and more nuanced conclusions regarding the support for a particular predictor driving trait evolution across a phylogeny.”

In the paper

“The former (hypothesis testing) seeks to understand the causes of evolutionary shifts, while the latter (data driven) is a descriptive and exploratory approach to understanding evolutionary patterns.”

In the paper

“Note that these scenarios do not differ qualitatively from Felsenstein's worst-case scenario nor the Phrynosoma-only model scenario from Case Studies I and II (Fig. 1).”

In the paper

“In the case of the PE hypothesis, we show that it does indeed have explanatory power beyond simply explaining a singular shift in Phrynosoma and support the original authors' conclusions.”

In the paper

“If so, then any number of causal factors that differ between Phrynosoma and other lizards could be equally as likely as PE-including FM with a slight reclassification of character states!”

In the paper

“While today we have a much wider range of comparative models to choose from including some that allow for adaptive shifts, most continuous trait models are Gaussian (e.g., Pagel 1999;Blomberg et al. 2003;Butler and King 2004;O'Meara et al. 2006;Eastman et al. 2011;Beaulieu et al. 2012;Uyeda and Harmon 2014) and do not accommodate abrupt, discontinuous shifts in phenotypes.”

What the sources say

• ?Not checked — no full text of the sourcecites Pagel 1999 source
• ?Not checked — no full text of the sourcecites Blomberg et al. 2003 source
• ?Not checked — no full text of the sourcecites Butler and King 2004 source

In the paper

“In all three scenarios, something novel and interesting happened on a single branch and the distribution of traits at the tips of the phylogeny reflects this.”

In the paper

“The resolution of Simpson's paradox rests entirely on causal assumptions; which are immediately apparent from graphical models but difficult to express with standard mathematical formulae.”

• Also stated in: § CONCLUDING REMARKS: ARE OUR MODELS VALID · plain

  “The resolution of Simpson's paradox rests entirely on causal assumptions; which are immediately apparent from graphical models but difficult to express with standard mathematical formulae.”

In the paper

“We also note that all the phylogenetic signal is coming from the evolution of body size, which becomes irrelevant once we observe body size, and thus we do not need to perform PGLS.”

In the paper

“One way to gain a foothold on the problem of causation is to build, communicate, and analyze PCMs in a graphical modeling framework-a perspective that has recently been advocated by Höhna et al. (2014Höhna et al. ( , 2016)).”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites Höhna et al. (2014, 2016)

In the paper

“In Case Study II, we present one solution to the problem that involves explicitly accounting for the possibility of unaccounted adaptive shifts using Bayesian Mixture modeling.”

In the paper

“In our example, a four species phylogeny provides possible pathways for causal effects, but variables may have entirely nonphylogenetic causes or may be blocked from ancestral causes by observed measurements, rendering the phylogeny irrelevant (e.g., Fig. 7A).”

In the paper

“The remarkable coincidence is not just that the two characters happened to evolve on the same branch but that they were never subsequently gained or lost throughout the rest of the tree.”

In the paper

“We have shown that discontinuous processes and singular events are poorly handled in our current framework and lead to much confusion about what exactly, our statistical methods are allowing us to infer from comparative data.”

In the paper

“As we illustrate through our Case Studies, we think that accounting for the possibility of idiosyncratic evolutionary events will be an essential step towards such a solution.”

In the paper

“This is in large part because comparative data and the phylogeny that underly it are largely beyond the control of the researcher, but careful consideration of the data is required to understand what research questions can be reasonably answered.”

In the paper

“For example, it is common practice among systematists to search for defining and diagnostic characteristics for named clades; these traits are of especial interest and are likely the same sorts of traits that are researchers might include in comparative analysis, thereby greatly increasing the likelihood of finding traits with independent, unrelated origins that align with Darwin's scenario.”

In the paper

“And it is clear from reading the macroevolutionary literature, biologists do not shy away from forming causal statements from correlative data regardless.”

In the paper

“but we do think that any compelling solution will involve a unification of phylogenetic natural history and hypothesis-testing approaches.”

In the paper

“We ultimately disagree with them that this constitutes a weak conclusion; the challenges of making these inferences from any comparative data set are significant.”

What the sources say

• ?Not checked — no full text of the sourcecites Maddison and FitzJohn (2015) source

In the paper

“However, such intermediate cases will only introduce slight differences and may not be distinguishable from the fully dependent case under Darwin's Scenario (though they will be important in more intermediate cases, see Supplementary Material available on Dryad).”

In the paper

“Understanding the exact mathematical reasons why Pagel's test infers a significant correlation in a given case provides a clear boundary condition that can help develop quantitative corrections for ascertainment bias.”

In the paper

“Therefore, we are assured that increasing body size will likewise always decrease species abundance.”

In the paper

“However, under network (B) the correct method is PGLS (gray line) as increasing body size will have no effect on migratory behavior, and unaccounted phylogenetic residual error is present in the observed data.”

In the paper

“By explicitly including phylogeny into our graphical models of causation, we are forced to reckon with the scope of the inference problem and the ability of our data to be informative.”

In the paper

“This is because increasing body size will simultaneously decrease species abundance and increase migratory behavior, which itself increases abundance, leading to a net slight increase in abundance.”

In the paper

“By making our models explicit, we see that the phylogeny is best thought of as a pathway for past factors to causally influence the present-day distribution of observed states.”

In the paper

“As a result, extending our analytical approach to more complicated scenarios will likely provide an even more meaningful estimate of the weight of evidence supporting a hypothesis of correlation.”

In the paper

“Maddison and FitzJohn (2015) hinted that the coincident occurrence of single events could be a way of measuring the evidence for a correlation, but did not work out the details as we have done here.”

What the sources say

• ?Not checked — no full text of the sourcecites Maddison and FitzJohn (2015) source

In the paper

“Indeed, we find that in all three Case Studies, the uniting philosophy is to consider models that account for background shifts in evolutionary regime, rather than strict adherence to a particular methodology.”

In the paper

“Furthermore, the issues of ascertainment bias are likely to rapidly dissipate as we move away from the boundary case of Darwin's scenario.”

In the paper

“However, from our example of bird size and abundance, it is apparent that Westoby et al. were right all along: PCMs are a powerful tools for drawing inferences from interspecific data but they necessarily imply some types of causal structures and negate others.”

In the paper

“Westoby and colleagues were concerned that including phylogeny in interspecific comparisons necessarily favored some causal explanations over others.”

What the sources say

• ?Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  cites Westoby et al. (1995a,b)