Computational methods for rapid structural modelling of antigen-antibody interactions to improve identification of antigen-specific antibodies from Ig

抗体精致的抗原识别特异性使其成为诊断、研究试剂和最成功的生物制药类别的有用材料。发现更好的基于抗体的治疗方法的能力需要了解整个抗体库中单个抗体的3D形状。下一代测序方法(Ig-seq)可以快速产生数百万个抗体基因序列。然而，到目前为止，无法常规地将抗体结构覆盖在大的Ig-seq数据集上，限制了它们发现抗体药物的潜力。在这个项目中，我们将使用计算方法在两个领域之间架起一座桥梁，允许对Ig-seq实验进行结构性注释，这将为更先进的基于抗体的疗法铺平道路。该项目属于MRC交付计划的范围，既涉及到优先挑战(例如，在血清学和疫情控制方面的应用，也包括在疫苗开发方面，后者将通过与牛津疫苗集团的合作以及加速发现基于抗体的疗法来实现)，以及它与MRC技术优先事项的一致性。相关的MRC技能优先考虑的是定量技能和跨学科技能。数学、统计学和计算的关键量化技能是蛋白质结构预测项目的核心。结构预测本身是跨学科的，借鉴了化学、统计学和生命科学的见解。此外，该项目的目的是将这些通过计算产生的结构模型与作为MRC信息学研究焦点的那种大规模和快速积累的序列数据联系起来。信息学和计算已被确定为MRC转变健康研究战略的关键步骤，因此，通过计算将高通量序列数据与结构生物学联系起来的项目恰好符合MRC的生物医学信息学倡议。该项目由Kymab共同监督和资助。

The exquisite antigen recognition specificity of antibodies has made them useful as diagnostics, research agents and the most successful class of biopharmaceuticals. The ability to discover better antibody-based therapeutics needs knowledge of the 3D shape of individual antibodies within the context of the entire antibody repertoire. Next-generation sequencing methodologies (Ig-seq) can rapidly yield millions of antibody gene sequences. However, so far, the inability to routinely overlay antibody structure on large Ig-seq datasets has limited their potential for antibody drug discovery. In this project we will use computational methodologies to bridge between the two fields by allowing structural annotation of Ig-seq experiments which will pave the way for more advanced antibody-based therapeutics.The project falls within the remit of the MRC delivery plan in both its connection to priority challenges (e.g. applications in serology and thus both outbreak control and in vaccine development, the latter of which will be realised through working with Oxford Vaccine Group, as well as accelerating discovery of antibody-based therapeutics) and its alignment with MRC skill priorities. The relevant MRC skill priorities are quantitative and interdisciplinary skills. The key quantitative skills of mathematics, statistics and computation are central to the project in the form of protein structure prediction. Structure prediction itself is interdisciplinary, drawing on chemistry, statistics and insight from the life sciences. Furthermore, the aim of the project is to connect these structural models, produced computationally, with the kind of large-scale and rapidly accruing sequence data that is at the focus of the MRC's research spotlight on informatics. Informatics and computation have been identified as key steps in the MRC's strategy for transforming health research, and thus the project, in linking high throughput sequence data to structural biology via computation, fits neatly into the MRC's initiative for biomedical informatics. The project is co-supervised and funded by Kymab.