Assessing the prediction fidelity of ancestral reconstruction by a library approach

Abstract

Ancestral reconstruction is a powerful tool for studying protein evolution as well as for protein design and engineering. However, in many positions alternative predictions with relatively high marginal probabilities exist, and thus the prediction comprises an ensemble of near-ancestor sequences that relate to the historical ancestor. The ancestral phenotype should therefore be explored for the entire ensemble, rather than for the sequence comprising the most probable amino acid at all positions [the most probable ancestor (mpa)]. To this end, we constructed libraries that sample ensembles of near-ancestor sequences. Specifically, we identified positions where alternatively predicted amino acids are likely to affect the ancestor's structure and/or function. Using the serum paraoxonases (PONs) enzyme family as a test case, we constructed libraries that combinatorially sample these alternatives. We next characterized these libraries, reflecting the vertebrate and mammalian PON ancestors. We found that the mpa of vertebrate PONs represented only one out of many different enzymatic phenotypes displayed by its ensemble. The mammalian ancestral library, however, exhibited a homogeneous phenotype that was well represented by the mpa. Our library design strategy that samples near-ancestor ensembles at potentially critical positions therefore provides a systematic way of examining the robustness of inferred ancestral phenotypes.

The molecular origin and evolution of dim-light vision in mammals

“…; Pollock and Chang ; Bar‐Rogovsky et al. ). Thus, we inferred three additional ancestral mammalian sequences, one by weighted random resampling from the posterior distribution and two by site‐directed mutagenesis of residues that were reconstructed with low posterior probability to an alternate residue in the distribution.…”

The thermostability and specificity of ancient proteins

“…For example, Hart et al measured ten alternate sequences of a ~3 billion year-old ancestor and found a T m of 76.7 ± 2 °C (compared to 68.0 °C of RNase H from E. coli ) [12••]. Using such approaches, many sources of random error have been investigated: uncertain tree topology [3,7,21,22], alternate evolutionary models [23], choice of reconstruction method [6,22], different amino acid frequencies [3], and reconstruction ambiguity [3,7,11,12••,24]. In all such studies, the properties of the ancestors have proven robust to uncertainty.…”

Experimental Evolution of Escherichia coli Harboring an Ancient Translation Protein

“…Ancestral sequence reconstruction may provide a means of addressing this limitation of the fossil record; the technique permits phylogenetics-based sequence inferences of ancestral genes at the interior nodes of a tree using likelihood or Bayesian statistics and offers an opportunity to determine the selectively advantageous amino acid replacements responsible for changes in protein behavior associated with adaptive events for particular molecular systems (Benner 1995; Chang et al 2002; Huelsenbeck and Bollback 2001; Liberles 2007; Pauling and Zuckerkandl 1963; Thornton 2004; Ugalde et al 2004). Mathematical sequence reconstructions of ancient genes and their subsequent in vitro biochemical characterization alone, however, may not necessarily provide the salient details of why the protein evolved along a particular evolutionary pathway (Bar-Rogovsky et al 2015; Copley 2012; Dean and Thornton 2007; Kacar 2016; Zhu et al 2005). Incorporating a functional perspective into the study of ancient proteins was suggested to be instrumental for understanding historical adaptive pathways as well as bridging the evolution of protein-level function and the organism-level behavior, thus enabling predictions that connect inferred genotype to ancestral phenotype (Dean and Thornton 2007; Harms and Thornton 2013; Kacar and Gaucher 2013, 2012; Lunzer et al 2005; Zhu et al 2005).…”