Tool for Predicting Glycosaminoglycan Recognition of Proteins

项目摘要 糖胺聚糖具有相当大的结构多样性，这使得对单个糖胺聚糖序列的研究成为可能 人类是不可能的。迄今为止进行的大多数研究依赖于异质性GAG组合物，如肝素 和硫酸软骨素。近年来，市场上已经有几十种GAG寡糖可供购买 （Sigma（美国）、Dextra（英国）和Iduron（英国））。然而，即使是购买一个小的，合理多样的图书馆， 低聚糖非常昂贵（每种几微克到毫克约200 -300美元）。更重要的是， 从这些公司可获得的序列通常是共同的序列， 在自然界中。GAG寡糖的合成具有挑战性，只有少数几个小组有经验， 合成技术我们已经开发了一种计算工具，可以帮助预测关键的GAG序列， 以高亲和力识别蛋白质。我们的工具已被验证的蛋白质，包括抗凝血酶，成纤维细胞 生长因子-1及其受体（FGF-1/FGFR 1）、转化生长因子β 2（TGF β 2）、凝血酶、组蛋白 乙酰转移酶p300、人嗜中性粒细胞弹性蛋白酶和趋化因子CXCL 13。我们建议让这个工具自由 可供研究界使用，以便许多小组可以通过计算评估他们的蛋白质是否 兴趣使GAG结合。我们的两个目标包括：1）在Web服务器上开发图形用户界面（GUI）， 使研究人员能够利用我们的计算工具来研究GAG-蛋白质相互作用; 2）推进 预测市售GAG序列（HP/HS）相互作用的计算工具 和CS/DS）与蛋白质。这两个目标通过使我们的内部工具“显著” 对非专业人士来说更直接、更容易理解”。在输出方面，这项工作将提出一个网络- 能够携带GAG序列库和适当算法的工具，供研究人员远程使用 网站.它将促进聚糖工具的持续民主化，以实现更有效的聚糖研究。在 就知识贡献而言，我们的计算工具将有助于加强对GAGs如何 被蛋白质识别，特别是那些属于凝血、炎症和生长/形态发生的蛋白质 系统.

PROJECT SUMMARY GAGs present considerable structural diversity, which has made the study of individual GAG sequences humanly impossible. Most studies performed to date rely on heterogeneous GAG compositions, such as heparin and chondroitin sulfate. Few dozen GAG oligosaccharides have become commercially available in recent times (Sigma (US), Dextra (UK), and Iduron (UK)). Yet, purchasing even a small, reasonably diverse library of these oligosaccharides is very expensive (~$200–300 for few µg to mg each). More importantly, the oligosaccharides available from these companies are generally the common sequences and do not represent the diversity present in nature. Synthesis of GAG oligosaccharides is challenging and only a handful of groups have experience with synthesis technology. We have developed a computational tool that helps predict key GAG sequence that recognize protein with high affinity. Our tool has been validated for proteins including antithrombin, fibroblast growth factor-1 & its receptor (FGF-1/FGFR1), transforming growth factor 2 (TGF2), thrombin, histone acetyltransferase p300, human neutrophil elastase and chemokine CXCL13. We propose to make this tool freely available to the research community so that many groups can computationally assess whether their protein of interest binds GAGs. Our two aims include 1) develop a graphical user interface (GUI) on a web-server to enable researchers utilize our computational tool for studying GAG–protein interactions; and 2) advance the computational tool for predicting the interaction of commercially available GAG sequences (HP/HS and CS/DS) with proteins. These two aims directly address the RFA by making our in-house tool “significantly more straightforward and accessible for non-specialists”. In terms of output, this work will put forward a web- enabled tool carrying libraries of GAG sequences and appropriate algorithms for use by researchers from remote sites. It will add to the continuing democratization of glycan tools to enable more effective glycan research. In terms of knowledge contribution, our computational tool would help enhance understanding on how GAGs are recognized by proteins, especially those belonging to coagulation, inflammation and growth/morphogenesis systems.