I like this question! When purifying a highly soluble, abundant, well behaved protein, all you need is good technical skills. Nevertheless, most interesting proteins are not that soluble, or abundant, or well behaved. In these situations you definitely need imagination and, even more importantly, experience allowing you to find the right combination among the hundreds of possibilities you have at your disposal. You need to understand your protein, you need to feel it, predict how it will behave, at the same time being aware that your prediction might be wrong.
References, if rich with details, are also very important, especially for challenging purification protocols, but they are just the beginning and you need experience, and sometimes imagination, to understand if you got all the details or something critical is missing.
Protein purification - and science in general - is both craft and art.
I like this question! When purifying a highly soluble, abundant, well behaved protein, all you need is good technical skills. Nevertheless, most interesting proteins are not that soluble, or abundant, or well behaved. In these situations you definitely need imagination and, even more importantly, experience allowing you to find the right combination among the hundreds of possibilities you have at your disposal. You need to understand your protein, you need to feel it, predict how it will behave, at the same time being aware that your prediction might be wrong.
References, if rich with details, are also very important, especially for challenging purification protocols, but they are just the beginning and you need experience, and sometimes imagination, to understand if you got all the details or something critical is missing.
Protein purification - and science in general - is both craft and art.
I agree with the excellent Carlo's comments and would like to add a little more. Before you develop your purification procedure you need to collect as much as possible information about your target protein: molecular weight, isoelectric point, membrane or cytoplasm location, pH optimum for bioactivity, substrate specificity, possible affinity ligands/substrates, potential conditions for inactivation, solubility, hydrophobicity, possible glycosylation, possible subunit structure, relative abundance in different sources, possible enzymatic assay to monitor the purification, and possible published protocols in literature for purification of this or similar proteins. The more information you collect and analyze the more efficient your purification procedure will be. The usual problem is that not much information might be available for new proteins, and in that case you need to use the info you have, your purification experience with other proteins, your inspiration, imagination, intuition, and small-scale pilot experiments results.
I think you need good understanding about purification principles and enzyme characteristics. Reference paper and your imagination can assist you once you are aware about purification principles and techniques.