Assessing Rapid Relaxed-Clock Methods for Phylogenomic Dating

Abstract

Rapid relaxed-clock dating methods are frequently applied to analyze phylogenomic data sets containing hundreds to thousands of sequences because of their accuracy and computational efficiency. However, the relative performance of different rapid dating methods is yet to be compared on the same data sets, and, thus, the power and pitfalls of selecting among these approaches remain unclear. We compared the accuracy, bias, and coverage probabilities of RelTime, treePL, and least-squares dating time estimates by applying them to analyze computer-simulated data sets in which evolutionary rates varied extensively among branches in the phylogeny. RelTime estimates were consistently more accurate than the other two, particularly when evolutionary rates were autocorrelated or shifted convergently among lineages. The 95% confidence intervals (CIs) around RelTime dates showed appropriate coverage probabilities (95% on average), but other methods produced rather low coverage probabilities because of overly narrow CIs of time estimates. Overall, RelTime appears to be a more efficient method for estimating divergence times for large phylogenies.

Accessing the relative performance of fast molecular dating methods for phylogenomic data

“…For most of these datasets, treePL placed the age of the deep nodes exactly at or very close to the values provided as loosse maxima. Additionally, our ndings corroborate Barba-Montoya et al (2021), where TreePL was more impacted by small deviations from the molecular clock. This probably resulted in more asymmetrical distributions of − D values for treePL, while RelTime presented lower asymmetry (Supporting information 2).…”

“…2 times wider than the Bayesian CIs. The greater precision of treePL when compared to Bayesian methods was also recovered using simulated data [47]. Simulations also have shown that RelTime CIs exhibit equivalent or greater coverage probabilities than Bayesian approaches [42].…”

“…Although it seems that fast-dating and Bayesian methods are asymptotically different, even in the dataset with the longest alignment length (Ran18), the inferred timescales led to very similar biological interpretation of the scenario in which lineage diversi cation took place. For RelTime, calibration density signi cantly impacted the MSE of time estimates, implying that, besides alignment length, increasing the number of time constraints also makes the differences between methods more pronounced [47].…”

“…As PL and RRF have been increasingly used to estimate timescales over the last years, it is important to carry out such largescale evaluation against the popular Bayesian framework. While previous studies investigated both fast dating methods separately [22,25,40,[43][44][45][46], a joint assessment of their performance is lacking on empirical data [47]. Moreover, treePL, which is the most popular implementation of PL for large phylogenies, was not extensively compared to any Bayesian method whatsoever, and there is little information on how they behave comparatively on real data.…”

Assessing the relative performance of fast molecular dating methods for phylogenomic data

“…As PL and RRF have been increasingly used to estimate timescales over the last years, it is essential to carry out large-scale evaluation against the popular Bayesian framework. While previous studies investigated both fast dating methods separately [ 22 , 25 , 40 , 45 – 48 ], a joint assessment of their performance with empirical data is lacking [ 49 ]. Moreover, treePL, which is the most popular implementation of PL for large phylogenies, was not extensively compared to any Bayesian method whatsoever, and there is little information on how they behave comparatively with real data.…”