问题：

蛋白质结构预测的一般过程。the general steps of Protein structur

---

---

答案：

首先，需要进行序列比对，在生物数据库中寻找与目标序列类似的已知序列。To begin with, we need to do sequence alignment to find similar sequence as the target sequence in bio-databases.第二，进一步确认，在数据库中，是否存在，与目标序列相类似的已知序列的三维结构信息，如果有，可以利用已知序列的蛋白质三维结构信息，进行同源比较建模，从而获得目标序列的三维结构预测。we need to find if the similar sequence has 3D structure information yet, if yes, we can use 3D comparative modeling and then finish the 3D structure Prediction of the target sequence.第三，如果在数据库中并没有与目标序列相类似的已知序列的三维结构信息，则需要做进一步的蛋白质家族，域，或聚类分析，以明确的目标序列与已知序列的结构信息之间的关系。if the answer is not, we need to do further Protein family, domain, or cluster analysis to clear the relationship between the target sequence and proteins with known structure information.第四，如果不能明确目标序列与相类似的已知序列结构信息之间的关系，或者现有数据库中并不存在与目标序列相似的三维结构信息，则需要采用头算分析法来预测目标序列的三维结构信息。If we cannot clear the relationship between the target sequence and proteins with known structure information, or we regret to find that there is no known structure similar to the target sequence, then we need to do Ab initio 3D analysis to predict the structure of the target sequence.