Empirically testing the accuracy and bias of ancestral protein resurrection methods

项目总结/摘要 祖先序列重建（ASR）方法现在已经广泛用于实验分析DNA序列。 古代生物分子的性质，并阐明分子进化的机制。通过重述 蛋白质在进化过程中的结构、机制和功能变化，ASR已经能够 解决许多传统方法无法解决的基本和具有挑战性的进化问题。 ASR方法在解决现代蛋白质的生物物理问题方面也非常成功， 近年来，它在蛋白质生物工程和设计中的实际应用引起了人们的注意。 尽管取得了这些进展，但复活的祖先序列的准确性、精确度和偏差目前仍是未知的。 未知最可能的祖先序列是否系统地偏向于具有异常的生物物理学特征 财产？复活的蛋白质的生化特性在多大程度上概括了 真正的祖先生物分子哪些进化模型提供了最准确的祖先重建？ 举一个众所周知的潜在偏见的例子，祖先复活的蛋白质通常比 热稳定性比他们的现代后代，目前有争议的是，是否观察到的高 热稳定性是古老蛋白质的真实性质，或者更确切地说是方法学上的人工产物。这些 这些问题很难明确回答，因为真实的祖先蛋白质通常会丢失， 历史，但我们的目标是提供实验性的答案。本提案将为（1）开发实验方法。 评估祖先复活的系统偏差，（2）评估重建蛋白质的准确性 通过与实际蛋白质的性质进行比较，（3）验证竞争性进化 ASR分析模型，以及（4）进化的祖先后验分布的高通量分析 研究和蛋白质工程。

Project Summary/Abstract Ancestral sequence reconstruction (ASR) methods have now become widely used to experimentally analyze the properties of ancient biomolecules and to elucidate the mechanisms of molecular evolution. By recapitulating the structural, mechanistic, and functional changes of proteins during their evolution, ASR has been able to address many fundamental and challenging evolutionary questions where more traditional methods have failed. ASR methodology has also been highly successful in addressing biophysical problems in modern proteins, and it has recently drawn attention for its practical applications in protein bioengineering and design. Despite these advances, the accuracy, precision, and bias of resurrected ancestral sequences is currently unknown. Are the most probable ancestral sequences systematically biased to have anomalous biophysical properties? How well do the biochemical properties of resurrected proteins recapitulate the properties of the true ancestral biomolecules? Which evolutionary models provide the most accurate ancestral reconstructions? To give one well-known example of a potential bias, ancestrally resurrected proteins are often much more thermostable than their modern descendants, and it is currently controversial whether the observed high thermostability is a bona fide property of ancient proteins or rather is a methodological artifact. These questions are extremely difficult to answer definitively because the real ancestral proteins are generally lost to history, but we aim to provide experimental answers. This proposal will develop experimental methods for (1) evaluating the systematic bias in ancestral resurrections, (2) assessing the accuracy of reconstructed protein properties by comparison with the properties of the actual proteins, (3) validating competing evolutionary models for ASR analyses, and (4) high throughput analysis of ancestral posterior distributions for evolutionary studies and protein engineering.