Bioinformatic amino acid sequence searches are used, in part, to assess the potential allergenic risk of newly expressed proteins in genetically engineered crops. Previous work has demonstrated that the searches required by government regulatory agencies falsely implicate many proteins from rarely allergenic crops as an allergenic risk. However, many proteins are found in crops at concentrations that may be insufficient to cause allergy. Here we used a recently developed set of high-abundance non-allergenic proteins to determine the false-positive rates for several algorithms required by regulatory bodies, and also for an alternative 1:1 FASTA approach previously found to be equally sensitive to the official sliding-window method, but far more selective. The current investigation confirms these earlier findings while addressing dietary exposure.

中文翻译：

---

使用官方生物信息学算法进行过敏原错误检测。

生物信息学氨基酸序列搜索部分用于评估基因工程作物中新表达的蛋白质的潜在致敏风险。先前的工作表明，政府监管机构进行的搜索错误地暗示了来自极少数致敏作物的许多蛋白质具有致敏风险。但是，在农作物中发现许多蛋白质，其浓度可能不足以引起过敏。在这里，我们使用了一组最新开发的高丰度非过敏性蛋白质，来确定监管机构所需的几种算法的假阳性率，以及先前发现对官方滑动同样敏感的另一种1：1 FASTA方法-window方法，但选择性更高。当前的研究证实了这些早期发现同时解决了饮食接触问题。