A Functional Census of p53 Cancer and Suppressor Mutants

描述（由申请人提供）：广泛的长期目标是：（1）证明计算和实验方法的合作，以实现具有重要医学意义的大突变序列空间的功能普查;（2）有助于我们对p53功能拯救机制的了解，从而促进对影响p53类似功能拯救的小分子癌症药物的研究;和（3）通过表征跨已知下游p53 DNA结合位点的p53癌症和抑制突变体的谱，阐明癌症的部分系统生物学。肿瘤抑制蛋白p53的突变发生在大约一半的人类癌症中，恢复突变缺陷的p53蛋白的功能是一个长期的医学目标。挽救p53癌症突变的生物学优先级在第二位点p53癌症抑制突变中发现。类似的p53药理学拯救每年将挽救数十万人的生命。理解和预测p53的拯救是实现这一目标的重要一步。 具体的目标是（1）计算预测p53癌症突变体的所有单一抑制突变，并通过实验验证结果，（2）通过两个或更多个协调突变的变化，优化已知和推定的p53抑制区的拯救效果，（3）预测和实验验证p53癌症和抑制突变体的DNA结合特异性已知的p53 DNA结合位点。我们的总体战略是一个协调的计算和实验攻击。我们已经有了实验性的p53功能测定和p53活性的计算预测，作为我们初步研究的一部分。计算预测器将用于将实验工作集中在最优先的领域。预测的实验验证将为机器学习技术带来更大的训练集。更大的训练集将导致更准确的预测，从而更集中的实验。因此，随着项目的进展，计算和实验之间的相互作用将变得更加有效。这一基本策略的变体适用于我们的每一个特定目标，这些目标都依赖于密切协调的实验和计算。

DESCRIPTION (provided by applicant): The broad, long-term objectives are (1) demonstrate computational and experimental methods cooperating to achieve a functional census of a large mutation sequence space of great medical importance; (2) contribute to our knowledge of p53 functional rescue mechanisms, and so facilitate the search for a small molecule cancer drug that effects an analogous functional rescue of p53; and (3) elucidate part of the systems biology of cancer by characterizing the spectrum of p53 cancer and suppressor mutants across known downstream p53 DNA binding sites. Mutations to the tumor suppressor protein p53 occur in approximately half of all human cancers, and restoring function to a mutationally defective p53 protein is a long-held medical goal. Biological precedence for rescuing p53 cancer mutations is found in second-site p53 cancer suppressor mutations. The analogous p53 pharmacological rescue would save hundreds of thousands of lives annually. Understanding and predicting p53 rescue is an important step toward that goal. The specific aims are (1) computationally predict all single suppressor mutations for p53 cancer mutants and validate the results experimentally, (2) optimize the rescue effects of known and putative p53 suppressor regions through two or more coordinated mutation changes, and (3) predict and experimentally validate the DNA binding specificity of p53 cancer and suppressor mutants for known p53 DNA binding sites. Our broad strategy is a coordinated computational and experimental attack. We already have experimental p53 functional assays and computational predictors of p53 activity, developed as part of our Preliminary Studies. Computational predictors will be used to focus experimental work into the highest priority areas. Experimental validation of the predictions will lead to a larger training set for machine learning techniques. The larger training set will lead to even more accurate predictions, thus even more focused experimentation. Thus, the interplay between computation and experiment will become ever more efficient as the project progresses. Variations of this basic strategy apply to each of our Specific Aims, which all rely on closely coordinated experiment and computation.