Improving prediction and detection of small proteins for functional analysis in a consortium resembling the functions of the human microbiome

尽管基因组学和蛋白质组学在过去几十年中取得了进展，但关于小蛋白的基因组和蛋白质组检测仍然是一个盲点。因此，人们对它们在原核生物中的功能知之甚少。从已知的例子中可以清楚地看出，s蛋白具有广泛的功能，其中一些被怀疑在塑造微生物群落方面也起到了作用。然而，这是微生物生命的自然形式，对sProteins在不同物种之间的相互作用和竞争中的作用知之甚少。在这项提案中，我们将调查一个类似于结肠微生物区系相关功能的模型群落。在这个群落中的8个物种中，我们将重点关注大肠杆菌和植物乳杆菌。为了提高对大肠杆菌和植物乳杆菌s蛋白的基因组识别能力，我们提出了对RNAcode的改进，这是一种识别正在进行纯化选择的多肽序列的方法，以达到足够的灵敏度和特异度来全面注释细菌基因组中的s蛋白质。这将通过采用更详细的统计模型来实现，用明确的概率计算来取代模拟，更好地处理对准误差。此外，我们将系统地处理功能肽仅是输入序列的子集保守的情况。该方法将作为综合数据分析管道提供。作为基于质谱学的检测的先决条件，丰富的s蛋白质必须通过特定的方法进行富集化。在这里，我们提出了微珠的应用，这些微珠在壳层显示出尺寸排斥功能，并在微珠的核心显示不同的层析特征。这些蛋白质将通过全球蛋白质组学和靶向蛋白质组学进行检测。随后，将根据序列特征和二级结构分析选定的一组s蛋白质的功能。利用圆二色谱和傅里叶红外光谱对合成的多肽进行二级结构测定。此外，经过生物素标记和亲和纯化后，将确定潜在的相互作用伙伴。生物信息学和蛋白质组学方法之间的密切互动将使我们能够改进细菌中小蛋白的预测、浓缩、检测和功能验证方面的工作。因此，我们的目标是识别新的s蛋白质以及它们如何塑造微生物群落的机制。

Despite the progress in genomics and proteomics over the last decades there is still a blind spot regarding the genomic and proteomic detection of small proteins (sproteins). As a consequence little is known about their functions in prokaryota. From the already known examples it becomes clear that sproteins fulfill a wide variety of functions, some of which are suspected to play a role also in shaping microbial communities. This, however, is the natural form of microbial life and very little is known abouth the role of sProteins in interaction and competition of different species. In this proposal we will investigate a model community resembling the relevant functionalities of the colon microbiota. Among the eight species in this community we will focus on E.coli and L. plantarum.In order to improve the genomic identification of sproteins for E.coli and L. plantarum we propose the improve to RNAcode, a method to identify peptide sequences that are under purifying selection, to reach sufficient sensitivity and specificity for a comprehensive annotation of sproteins in bacterial genomes. This will be achieved by employing a more elaborate statistical model, the replacement of simulations with explicit computations of probabilities, a better handling of alignment errors. Furthermore we will systematically handle the case that the functional peptide is conserved only is a subset of the input sequences. The method will be made available as an integrated data analysis pipeline. As a prerequisite for the mass spectrometry based detection the abundant sproteins have to be enriched by specific protocols. Here we propose the applications of microbeads that exhibit a size exclusion functionality at the shell and different chromatographic features in the core of the beads. The sproteins will be detected either by global as well as targeted proteomics. Subsequently the functions of a selected set of sproteins will be analyzed in terms of sequence features and secondary structure. The secondary structure will be determined from synthesized peptides by Circular Dichroism and Fourier infrared spectroscopy. In addition potential interaction partners will be identified after biotin-tagging and affinity purification. The close interaction between the bioinformatic and proteomic approach will allow us to improve the aspects of prediction, enrichment, detection and functional validation of small proteins in bacteria. Hence we are aiming at identifying novel sproteins and the mechanisms how they shape microbial communities.