APPLICATIONS OF ARTIFICIAL INTELLIGENCE FOR CHEMICAL INFERENCE .22. AUTOMATIC RULE FORMATION IN MASS-SPECTROMETRY BY MEANS OF META-DENDRAL PROGRAM
APPLICATIONS OF ARTIFICIAL INTELLIGENCE FOR CHEMICAL INFERENCE .22. AUTOMATIC RULE FORMATION IN MASS-SPECTROMETRY BY MEANS OF META-DENDRAL PROGRAM
复制标题
DOI：
10.1021/ja00436a017
复制
发表时间：
1976-01-01
影响因子：
15
通讯作者：
DJERASSI, C
中科院分区：
文献类型：
--
作者：
BUCHANAN, BG;SMITH, DH;DJERASSI, C
DENDRAL 计算机程序使用既定的分子碎片规则来帮助化学家从质谱数据中解决复杂的结构问题。描述了一种称为 Meta-DENDRAL 的计算机程序，它可以帮助从已知化合物的经验数据中发现此类规则。该程序使用启发式方法来搜索这些债券周围的常见结构环境，这些债券被发现会碎片化并抽象出合理的碎片规则。该程序在脂肪胺的充分表征的低分辨率质谱和雌激素类固醇的高分辨率质谱上进行了测试。该计划发现了单酮、二酮和三酮雄甾烷的新碎片规则。
The DENDRAL computer program uses established rules of molecular fragmentation to help chemists solve complex structural problems from mass spectral data. A computer program called Meta-DENDRAL, that can aid in the discovery of such rules from empirical data on known compounds, is described. The program uses heuristic methods to search for common structural environments around those bonds that are found to fragment and abstracts plausible fragmentation rules. The program was tested on the well-characterized, low-resolution mass spectra of aliphatic amines and the high-resolution mass spectra of estrogenic steroids. The program has discovered new fragmentation rules for mono-, di- and triketoandrostanes.