基于大数据的中国人抗体组库量化新方法

Antibody repertoire usually refers to the entire set of antibody heavy and light chain variable sequences captured via next generation sequencing from a blood or tissue sample. It is a rapidly developing approach in B cell immune response study which has led to the discovery of antibody binding to viruses and cancer neoantigens. It is also proved valuable in evaluating antigenicity after vaccine and may shed light to the mechanism of transplant rejection and autoimmune diseases. The antibodies binding to current antigen usually outnumbers others through a biological process called clonal expansion. Along with the rapid dividing of B cells, mutations are introduced into the variable regions of heavy and light chain sequences to enhance their antigen binding affinity. Thus the key to antibody repertoire studies is to acquire the intact heavy and light chain variable regions with their respective abundance. PCR amplification is an essential step for NGS library preparation. B cell genome has only one copy of heavy and light chain sequences each, hence the antibodies of interests are easily missed. cDNA library, which contains multiple copies of heavy and light chain sequences, is a preferable starting material. During amplification, although 5’RACE has no amplification bias, the resultant sequence length is beyond the current NGS limit. Therefore despite the amplification bias, multiplex PCR with 5’ primer pool is widely adopted in the field. Understandably, when Nussenzweig lab designed a new set of mixed primers based on the leader region rather than the highly mutable V gene segments, much more variable region sequences were captured. Researchers have used this primer set extensively to study antibody repertoire for the last few years. However, after scrutinized all leader regions on IMGT(the international ImMunoGeneTics information system for immunoglobulins or antibodies) we found this widely used primer set is flawed. Firstly, there are no primers designed to capture antibodies generated from V gene segments in the 2nd family. This may miss the antibodies that play critical roles in immune response. Secondly, a few pairs of the primers may target the same antibodies and cause nested PCR effect. Moreover, we analyzed the leader region sequences from five available genomes and ten datasets from Chinese people and surprisingly found that leader region is highly diversed among people from different regions. The primer set designed by Nussenzweig lab does not work in Chinese people. Based on the aforementioned evidence, we propose to redesign a primer set specifically for Chinese people. We plan to capture the antibody repertoire from 100 Chinese people using unbiased 5’RACE and systematically extract the leader regions. Using our published SONAR bioinformatics pipeline to acquire the frequently used V gene segments. We will then systematically sample all possible oligonucleotides with proper length for PCR primers and choose the primer combination with least number of sequences that can canvas all possible antibodies. Based on the selected primer set, we will design spike in sequences and corresponding bioinformatics analysis pipeline to remove the intrinsic PCR bias. Thus provide researchers a best primer set and bioinformatics analysis tool both capture all possible antibodies and acquire precise abundance of each antibody when study Chinese antibody repertoire.

B细胞通过产生抗体影响抗感染、疫苗接种、自身免疫疾病和肿瘤等，高度多样性的抗体组成的抗体组库不仅反映机体的免疫状况，也极大地影响疾病进程。基于高通量测序的抗体组库检测已经广泛应用于基础研究和临床实践并取得了显著成果。目前抗体组库测序主要分为基因组DNA扩增及重排后cDNA扩增两种方法，后者因具有真实反映抗体编码特征和高拷贝数的优势被广泛采用。然而现有的cDNA扩增方法存在两个重要的缺陷：1）由于多重引物的扩增偏差导致不能准确定量分析抗体拷贝数；2）目前的引物设计基于西方人群抗体序列特征， 因而在实践中不适合中国群体的分析。为此，本研究首先对中国人抗体序列的特异性进行大规模调查，然后根据这些序列特征设计适合于中国群体抗体组库检测的引物池和分析方法，并用于研究中国健康群体抗体组库的特征。本课题的成功实施将对基于抗体组库的诊断和抗体药物开发研究产生重要的指导意义。

B细胞作为体液免疫的重要成员，其表面受体BCR在免疫应答过程中发挥极其重要的作用。随着高通量测序的发展，获得性免疫组库测序技术（AIRR）使得研究者可一次性获得数以百万计的抗体序列。尽管AIRR已在疫苗设计、抗体研发、发病机制探究等方面展现了巨大的潜力，但是仍存在很多棘手的问题。例如，胚系参考序列的不完整性导致研究者无法捕获更为全面的抗体组库；扩增过程中产生的错误，比如嵌合体序列等，影响了抗体组库的准确定量。本项目拟：1）借助RACE扩增方式获得的数据，补充完善人类、恒河猴以及大小鼠胚系参考序列；2）通过添加分子标识符的方式，改进现有实验流程，实现准确的抗体组库定量；3）比较现有分析工具，选择性能更好的软件，搭建标准化免疫组库测序数据分析流程；4）从大数据角度出发，描述免疫组库图谱；5）整合抗体组库大数据，搭建网络分析平台。