BACKGROUND & AIMS: :Cytochrome P450 enzymes are responsible for phase I metabolism of the majority of drugs and xenobiotics. Identification of the substrates and inhibitors of these enzymes is important for the analysis of drug metabolism, prediction of drug-drug interactions and drug toxicity, and the design of drugs that modulate cytochrome P450 mediated metabolism. The substrates and inhibitors of these enzymes are structurally diverse. It is thus desirable to explore methods capable of predicting compounds of diverse structures without over-fitting. Support vector machine is an attractive method with these qualities, which has been employed for predicting the substrates and inhibitors of several cytochrome P450 isoenzymes as well as compounds of various other pharmacodynamic, pharmacokinetic, and toxicological properties. This article introduces the methodology, evaluates the performance, and discusses the underlying difficulties and future prospects of the application of support vector machines to in silico prediction of cytochrome P450 substrates and inhibitors.

背景与目标： ：细胞色素P450酶负责大多数药物和异种生物的I期代谢。这些酶的底物和抑制剂的鉴定对于分析药物代谢，预测药物相互作用和药物毒性以及设计可调节细胞色素P450介导的代谢的药物非常重要。这些酶的底物和抑制剂在结构上是多种多样的。因此，期望探索能够预测多种结构的化合物而不会过度拟合的方法。支持向量机是具有这些特性的一种有吸引力的方法，已被用于预测几种细胞色素P450同工酶以及各种其他药效，药代动力学和毒理学性质的化合物的底物和抑制剂。本文介绍了该方法，评估了性能，并讨论了支持向量机在计算机上预测细胞色素P450底物和抑制剂的计算机应用中的潜在困难和未来前景。

BACKGROUND & AIMS: MyristoylCoA:protein N-myristoyltransferase (NMT) catalyzes the cotranslational covalent attachment of a rare cellular fatty acid, myristate, to the N-terminal Gly residue of a variety of eukaryotic proteins. The myristoyl moiety is often essential for expression of the biological functions for these proteins.

Attachment of C14:0 alone provides barely enough hydrophobicity to allow stable association with membranes. The partitioning of N-myrisotylproteins is therefore often modulated by "switches" that function through additional covalent or noncovalent modifications. Candida albicans, the principal cause of systemic fungal infection in immunocompromised humans, contains a single NMT gene that is essential for its viability. The functional properties of the acylCoA binding site of human and C. albicans NMT are very similar. However, there are distinct differences in their peptide binding sites. An ADP ribosylation factor (Arf) is included among the few cellular protein substrates of the fungal enzyme. Alanine scanning mutagenesis of an octapeptide derived from an N-terminal Arf sequence (GLYASKLS-NH2) disclosed that Gly1, Ser5, and Lys6 play predominant roles in binding. ALYASKLS-NH2 is an inhibitor competitive for peptide [Ki(app) = 15.3 +/- 6.4 microM] and noncompetitive for myristoylCoA. Remarkably, replacement of the N-terminal tetrapeptide with an 11-aminoundecanoyl group results in a competitive inhibitor (11-aminoundecanoyl-SKLS-NH2) that is approximately 40-fold more potent [Ki(app) = 0.40 +/- 0.03 microM] than the starting octapeptide. Removal of Leu-Ser from the C-terminus generates a competitive dipeptide inhibitor (11-aminoundecanoyl-SK-NH2) with a Ki(app) of 11.7 +/- 0.4 microM, equivalent to that of the starting octapeptide. A derivative dipeptide inhibitor containing a C-terminal N-cyclohexylethyl lysinamide moiety has the advantage of being more potent (IC50 = 0.11 +/- 0.03 microM) and resistant to digestion by cellular carboxypeptidases. Rigidifying the flexible aminoundecanoyl chain results in very potent general NMT inhibitors (IC50 = 40-50 nM). Substituting a 2-methylimidazole for the N-terminal amine and adding a benzylic alpha-methyl group with R stereochemistry to the rigidifying element produces even more potent inhibitors (IC50 = 20-50 nM) that are up to 500-fold selective for the fungal compared to human enzyme. A related less potent member of this series of compounds is fungistatic. Its growth inhibitory effects are associated with a reduction in cellular protein N-myristoylation, judged using cellular Arf as a reporter. These studies establish that NMT is a new antifungal target.

背景与目标： MyristoylCoA ：蛋白N-肉豆蔻酰基转移酶（NMT）催化稀有细胞脂肪酸肉豆蔻酸酯与多种真核蛋白N-末端Gly残基的共翻译共价连接。肉豆蔻酰基部分通常对于表达这些蛋白质的生物学功能是必不可少的。

C14 ：0的附着本身仅提供了不足以使与膜稳定结合的疏水性。因此，N-肉豆蔻酰蛋白的分区通常由通过附加的共价或非共价修饰起作用的“开关”调节。白色念珠菌是免疫功能低下的人体内全身真菌感染的主要原因，它含有一个对其生存能力至关重要的单个NMT基因。人和白色念珠菌NMT的酰基辅酶A结合位点的功能特性非常相似。但是，它们的肽结合位点存在明显差异。 ADP核糖基化因子（Arf）包括在真菌酶的几种细胞蛋白底物中。来自N末端Arf序列（GLYASKLS-NH2）的八肽的丙氨酸扫描诱变显示，Gly1，Ser5和Lys6在结合中起主要作用。 ALYASKLS-NH2是一种对肽[Ki（app）= 15.3 /-6.4 microM]具有竞争性的抑制剂，对肉豆蔻酰辅酶A不具有竞争性。值得注意的是，用11-氨基十一烷酰基取代N末端四肽会产生竞争性抑制剂（11-氨基十一烷酰基-SKLS-NH2），其效力比[Ki（app）= 0.40 /-0.03 microM]高40倍左右。起始八肽。从C末端去除Leu-Ser会产生竞争性的二肽抑制剂（11-氨基十一烷酰基-SK-NH2），Ki（app）为11.7 /-0.4 microM，与起始八肽相当。含有C-末端N-环己基乙基赖氨酰胺部分的衍生物二肽抑制剂具有更有效的优势（IC 50 ＝ 0.11 /0.03μM），并且对细胞羧肽酶的消化具有抵抗力。刚性化柔性氨基十一烷酰基链会产生非常有效的常规NMT抑制剂（IC50 = 40-50 nM）。用2-甲基咪唑取代N端胺，并在刚性元素中添加具有R立体化学的苄基α-甲基，可产生更强效的抑制剂（IC50 = 20-50 nM），对真菌的选择性高达500倍与人类酶相比。该系列化合物的一个相关的效力较弱的成员是抑真菌的。使用细胞Arf作为报告基因判断，其生长抑制作用与细胞蛋白N-肉豆蔻酰化的减少有关。这些研究确定NMT是一种新的抗真菌靶标。