Characterizing the Effects of Protein and RNA Variability in Molecular Function and Interactions

提案摘要 该MIRA for ESI项目旨在研究和表征 生物分子，它与分子进化的联系，以及在科学和 生物医学应用它由两个科学目标组成，描绘了我们实验室的愿景。 我们的第一个目标是开发全球概率和计算模型，这将有助于我们回答 假设蛋白质变异性及其相互作用的景观可以表征和量化。 我们将通过基于从以下方面获得的大量数据创建概率模型来构建我们的框架： 测序，我们将对特定的效果进行详细的预测和实验验证。 我们的方法预测的突变。我们感兴趣的是功能突变的景观， 比破坏性突变空间更难描述。我们的第二个目标是扩大我们的 分子相互作用假说，包括核酸，特别是蛋白质-RNA相互作用。我们将 整合测序技术和计算方法来推断蛋白质突变景观， RNA识别我们将检验这样的假设，即不仅天然核酸基序可以被选择性地 识别，但也从我们的定量模型衍生的变体。我们将开发一个框架来编码 并从推断的景观中预测识别，并计划将我们的结果与实验结果相结合， RNA结合蛋白的技术可以帮助证实我们的假设。 在过去的几年里，我们的实验室已经能够推断蛋白质序列家族的全球模型，并量化 共同进化的信号从这些模型成功。这些全球模型在研究中产生了影响 蛋白质折叠，蛋白质动力学和蛋白质复合物的预测及其在 药物界面发现和药物-基因相互作用。生物分子的功能变体很难被 阐明，因为破坏性突变空间占主导地位。PI能够显示氨基酸的信号 或核酸共同进化也可以用作预测工具，以探索和编码功能性的 生物分子的突变空间。这个想法代表了一种范式转变， 信号可以用作发现机制。该项目的总体愿景旨在量化和揭示 由进化过程塑造的功能性生物分子可变性的光谱。这将帮助我们 与生物医学相关的发展，例如推断疾病突变的影响，抗生素耐药性， 生物分子传感器设计以及序列组成如何影响相互作用网络。

Proposal Summary This MIRA for ESI project proposes to investigate and characterize functional mutational variability in biomolecules, its connections to molecular evolution and the ability to use these landscapes in scientific and biomedical applications. It is composed of two scientific objectives that delineate the vision of our laboratory. Our first goal is to develop global probabilistic and computational models that will help us answer the hypothesis that the landscape of protein variability and their interactions can be characterized and quantified. We will build our framework by creating probabilistic models based on large quantities of data obtained from sequencing and we will make detailed predictions and experimental confirmation on the effect of specific mutations predicted by our methodology. We are interested in the landscape of functional mutations, which is much harder to characterize than the disruptive mutational space. For our second goal, we will expand our hypothesis to molecular interactions that include nucleic acids, particularly protein-RNA interactions. We will integrate sequencing technology and computational approaches to infer mutational landscapes of protein- RNA recognition. We will test the hypothesis that not only native nucleic acid motifs can be selectively recognized but also variants derived from our quantitative models. We will develop a framework to encode and predict recognition from inferred landscapes and plan to integrate our results with experimental technologies on RNA binding proteins that could help confirm our hypothesis. In the past few years our lab has been able to infer global models of families of protein sequences and quantify coevolutionary signals from these models successfully. These global models have had an impact in the study of protein folding, protein dynamics and the prediction of protein complexes as well as their applications in druggable interface discovery and drug-gene interactions. Functional variants of biomolecules are hard to elucidate, as the disruptive mutational space is dominant. The PI was able to show that signals of amino acid or nucleic acid coevolution can also be used as predictive tools to explore and encode the functional mutational space of biomolecules. This idea represents a paradigm shift where quantification of evolutionary signals can be used as a discovery mechanism. The overall vision of this project aims to quantify and uncover the spectrum of functional biomolecular variability sculpted by evolutionary processes. This will help us work on developments related to biomedicine, such as inferring the effects of mutations in disease, antibiotic resistance, biomolecular sensor design and how sequence composition has an impact on interaction networks.