Accurate Protein Identification using Peptide Separation Information and Tandem M

描述（由申请人提供）：目前阻碍蛋白质组学研究的一个重要问题是从质谱数据中鉴定肽和蛋白质。这个问题表现在难以建立蛋白质和肽鉴定的置信限。一个严重的问题是假阳性赋值的发生趋势。开发一个综合的蛋白质鉴定平台将大大降低这种分配的风险。这将使预测生理学和医学公司。(PPM)以产生人类血浆中蛋白质的无与伦比的特征。有了这种能力，PPM将能够进行必要的人口研究，以创建与健康和疾病相关的蛋白质图谱，称为可寻址数字阵列图谱（ADAM）。区分健康和疾病亚当斯的能力将大大增强PPM在个性化医疗领域的竞争力。此外，该平台本身也将是可销售的。 集成的蛋白质鉴定平台将使用通常在典型的液相色谱-质谱实验（LC-MS）中丢弃的信息创建。将开发旨在预测二维（2D）离子迁移谱（IMS，目前仅用于蛋白质组学研究的PPM技术）的肽分离曲线的算法。预测谱可用于减小蛋白质数据库大小，这减小了用于分配的同一性阈值;这有效地增加了肽分配的置信度以及增加了置信分配的数量。此外，将开发算法，以允许使用多个蛋白质数据库搜索程序。评分方案将根据最终作业的并集和交集制定。最后，算法和预测能力将结合在一个集成系统中，以确定输入亚当斯的蛋白质分配的准确p值。集成蛋白质识别平台的开发将对个性化医疗领域产生重大影响。这样的平台将能够开发前所未有的血浆蛋白数据库，可用于确定不同个体的疾病易感性和治疗反应。

DESCRIPTION (provided by applicant): A significant problem currently hindering proteomics studies is the identification of peptides and proteins from mass spectral data. This problem is manifest in the difficulty to establish confidence limits of protein and peptide identifications. A serious concern is the tendency for the occurrence of false positive assignments. The development of an integrated protein identification platform will greatly reduce the risk of such assignments. This will allow Predictive Physiology and Medicine, Inc. (PPM) to produce unrivaled characterizations of the proteins in human plasma. With this capability, PPM will be able to perform the population studies necessary to create the protein profiles associated with health and disease, termed Addressable Digital Array Maps (ADAM). The ability to distinguish healthy and disease ADAMs will greatly enhance the competitiveness of PPM in the field of personalized medicine. Additionally, the platform itself will be marketable. The integrated protein identification platform will be created using information that is generally discarded in typical liquid chromatography-mass spectrometry experiments (LC-MS). Algorithms will be developed aimed at predicting peptide separation profiles for two-dimensional (2D) ion mobility spectrometry (IMS, a technology currently only utilized by PPM for proteomics studies). The predicted profiles can be used to decrease protein database sizes which decrease the identity thresholds for assignments; effectively this increases the confidence of peptide assignments as well as increases the number of confident assignments. Additionally, algorithms will be developed to allow the use of multiple protein database search programs. A scoring scheme will be developed based on the union and intersection of resulting assignments. Finally, the algorithms and predictive capabilities will be combined in a single integrated system to determine an accurate p-value for protein assignments which are entered into the ADAMs.The development of the integrated protein identification platform will have a substantial impact on the field of personalized medicine. Such a platform will enable the development of unprecedented plasma protein databases that can be used to determine the disease susceptibility and therapeutic response of different individuals.