The structures of protein and DNA were discovered primarily by means of synthesizing component-level information about bond types, lengths, and angles, rather than analyzing X-ray diffraction photographs of these molecules. In this paper, I consider the synthetic and analytic approaches to exemplify alternative heuristics for approaching mid-twentieth-century macromolecular structure determination. I argue that the former was, all else being equal, likeliest to generate the correct structure in the shortest period of time. I begin by characterizing problem solving in these cases as proceeding via the elimination of candidate structures through the successive application of component-level information and interpretations of X-ray diffraction photographs, each of which serves as a kind of constraint on structure. Then, I argue that although each kind of constraint enables the elimination of a considerable proportion of candidate structures, component-level constraints are significantly more likely to do so correctly. Thus, considering them before X-ray diffraction photographs is a better heuristic than one that reverses this order. Because the synthetic approach that resulted in the determination of the protein and DNA structures exemplifies such a heuristic, its use can help account for these discoveries.

中文翻译：

---

蛋白质和 DNA 结构测定的合成与分析方法

蛋白质和 DNA 的结构主要是通过合成有关键类型、长度和角度的组分级信息来发现的，而不是分析这些分子的 X 射线衍射照片。在本文中，我考虑了综合和分析方法，以举例说明接近 20 世纪中叶大分子结构确定的替代启发式方法。我认为，在其他条件相同的情况下，前者最有可能在最短的时间内生成正确的结构。我首先将在这些情况下解决问题的特征描述为通过连续应用组件级信息和 X 射线衍射照片的解释来消除候选结构，每一个都是对结构的一种约束。然后，我认为，尽管每种约束都可以消除相当大比例的候选结构，但组件级约束更有可能正确地做到这一点。因此，在 X 射线衍射照片之前考虑它们是比颠倒这个顺序更好的启发式方法。因为导致确定蛋白质和 DNA 结构的合成方法体现了这种启发式方法，所以它的使用可以帮助解释这些发现。