The Protein Data Bank (PDB) includes a carefully curated treasury of experimentally derived structural data on biological macromolecules and their various complexes. Such information is fundamental for a multitude of projects that involve large-scale data mining and/or detailed evaluation of individual structures of importance to chemistry, biology and, most of all, to medicine, where it provides the foundation for structure-based drug discovery. However, despite extensive validation mechanisms, it is almost inevitable that among the ∼215 000 entries there will occasionally be suboptimal or incorrect structure models. It is thus vital to apply careful verification procedures to those segments of the PDB that are of direct medicinal interest. Here, such an analysis was carried out for crystallographic models of L-asparaginases, enzymes that include approved drugs for the treatment of certain types of leukemia. The focus was on the adherence of the atomic coordinates to the rules of stereochemistry and their agreement with the experimental electron-density maps. Whereas the current clinical application of L-asparaginases is limited to two bacterial proteins and their chemical modifications, the field of investigations of such enzymes has expanded tremendously in recent years with the discovery of three entirely different structural classes and with numerous reports, not always quite reliable, of the anticancer properties of L-asparaginases of different origins.

中文翻译：

---

为 L-天冬酰胺酶研究提供可靠的结构数据集。


蛋白质数据库 (PDB) 包括一个精心策划的宝库，其中包含生物大分子及其各种复合物的实验衍生结构数据。这些信息对于涉及大规模数据挖掘和/或对化学、生物学，最重要的是医学而言重要的个体结构的详细评估的众多项目来说是基础，它为基于结构的药物发现提供了基础。然而，尽管有广泛的验证机制，在〜215 000个条目中几乎不可避免地偶尔会出现次优或不正确的结构模型。因此，对 PDB 中具有直接医学意义的部分应用仔细的验证程序至关重要。在这里，对 L-天冬酰胺酶的晶体模型进行了这样的分析，这种酶包括已批准用于治疗某些类型白血病的药物。重点是原子坐标是否遵守立体化学规则以及它们与实验电子密度图的一致性。虽然 L-天冬酰胺酶目前的临床应用仅限于两​​种细菌蛋白及其化学修饰，但近年来，随着三种完全不同结构类别的发现，此类酶的研究领域已大大扩展，并有大量报告，但并不总是完全一致。不同来源的 L-天冬酰胺酶的抗癌特性的可靠性。