AbEMap results: Epitope Maps of AF2 folded (with template) Ab inputs

The third application of AbEmap is to use computationally modelled antibody structures to map epitopes on given antigen structures. This dataset contains epitope mapping results of 40 BM5 antigens when using AlphaFold folded antibody structures (Note - Antibody folding was guided by 1 template at a time). The resulting AbEMap scores obtained by using each AF model are averaged and are saved as b-factors in the PDB files included in this data set. eg. Antibodies in PDB IDs 1A6T, 2Z91, 3B9K, 3OPZ, 5KTE, 5KVG and 5TPW were used as templates when folding Antibody Sequence in 1AHW with AF2. The Top model from each run is then run through AbEMap and the final 1AHW's antigen epitope mapping results are the average of all six runs.

AbEmap的第三项应用为借助计算建模得到的抗体（antibody）结构，对指定抗原（antigen）结构上的表位（epitope）进行定位映射。本数据集收录了40种BM5抗原在使用阿尔法折叠（AlphaFold）预测的抗体结构进行表位映射所得到的结果（注：单次抗体折叠过程仅以单条模板作为指导）。将每一个阿尔法折叠模型所得到的AbEmap评分取平均值，并以B因子（b-factors）的形式保存于本数据集附带的蛋白质数据库（PDB）文件中。例如，在对1AHW对应的抗体序列使用阿尔法折叠2（AlphaFold 2）进行折叠预测时，选取了PDB编号为1A6T、2Z91、3B9K、3OPZ、5KTE、5KVG及5TPW的抗体作为折叠模板。随后将每一轮预测得到的最优模型输入AbEmap进行分析，最终1AHW的抗原表位映射结果为全部六轮实验结果的平均值。