摘要
机器人与机器学习的新闻编辑每日新闻-机器学习的新研究是一篇报道的主题。根据NewsRx记者在意大利Sesto Fiorenti NO的新闻报道，研究表明：“了解金属酶活性位点抑制剂结合的精细结构细节，可以对针对这类广泛B分子的合理药物设计产生深远的影响。结构技术如NMR、Cryo-EM和X-射线晶体学HY可以提供键长和角度。”但这些措施的不确定性可能与所有已发表的结构中观察到的这些条件的数值范围一样大。新闻记者从弗洛伦斯大学的研究中引用了一句话：“这种不确定性太大，无法在量子化学(QC)水平上进行可靠的计算，以开发精确的构效关系或改进蛋白质抑制剂研究中的能量考虑。因此，在最近的一篇论文中，我们已经证明，有可能精制钴(ii)取代的MMP12的活性位点，这是一种金属蛋白，它是一种相关的药物靶点。通过与多参考从头算QC方法计算的实验假接触位移S(PCS)相匹配,当起始结构与最终结构不够接近时,这种方法的计算成本成为一个重要的瓶颈。本文基于神经网络(NN)和支持向量回归(SVR)建立了一个APPROACH模型,并将其应用于另一个典型金属蛋白靶物草酸盐抑制的人碳酸酐酶2(hCAII)的活性位点结构的精化,精化后的结构与实验结果吻合良好。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Machine Learning is th e subject of a report. According to news reporting originating in Sesto Fiorenti no, Italy, by NewsRx journalists, research stated, “Understanding the fine struc tural details of inhibitor binding at the active site of metalloenzymes can have a profound impact on the rational drug design targeted to this broad class of b iomolecules. Structural techniques such as NMR, cryo-EM, and X-ray crystallograp hy can provide bond lengths and angles, but the uncertainties in these measureme nts can be as large as the range of values that have been observed for these qua ntities in all the published structures.” The news reporters obtained a quote from the research from the University of Flo rence, “This uncertainty is far too large to allow for reliable calculations at the quantum chemical (QC) levels for developing precise structure-activity relat ionships or for improving the energetic considerations in protein-inhibitor stud ies. Therefore, the need arises to rely upon computational methods to refine the active site structures well beyond the resolution obtained with routine applica tion of structural methods. In a recent paper, we have shown that it is possible to refine the active site of cobalt(II)-substituted MMP12, a metalloprotein tha t is a relevant drug target, by matching to the experimental pseudocontact shift s (PCS) those calculated using multireference ab initio QC methods. The computat ional cost of this methodology becomes a significant bottleneck when the startin g structure is not sufficiently close to the final one, which is often the case with biomolecular structures. To tackle this problem, we have developed an appro ach based on a neural network (NN) and a support vector regression (SVR) and app lied it to the refinement of the active site structure of oxalate-inhibited huma n carbonic anhydrase 2 (hCAII), another prototypical metalloprotein target. The refined structure gives a remarkably good agreement between the QC-calculated an d the experimental PCS.”
NETL
NSTL