A fairly senior scientists once gave me the advice that “you should never employ a lab technician who can’t cook”. A quantity of wine might have been consumed prior to this revelation but all the same I’m going to use it as an excuse to use lots of cooking analogies in this post. I hope it will help as I try to explain why protocols aren’t methods.
For the sake of this argument I’m going to classify research papers into three types. There are the vast array of Research articles which report the results of an experiment or experiments; there are Methods papers which present a new approach to collecting data and in which the reported results’ only importance is in showing that the approach actually works; and there are Protocols. Each contains a description of how experiments are performed but in very different levels of detail and for very different purposes.
Reading a research paper is like going to a restaurant. What you are really interested in is the results of someone else’s cookery: the food on your plate. You are interested to know how it has been produced to a degree – whether your meat is roasted or pan fried, are those potatoes dauphinoise or scallion mash, etc., – but you would have to already be an expert to use the information on the menu to recreate the meal. This is actually a real problem with research papers. Everything has to be taken on trust as the methods described are so scanty that there is no way that the results can be independently validated by replicating the experiments.
Taking things on trust is not science, but that is a rant for another day.
Reading a Methods papers is more like reading a glossy cookery book. I have loads of them on my bookshelves at home, many of them by celebrity chefs. They are great. They have lots of pictures of food and descriptions of where the chef discovered the recipes. Purple prose and travelogue into which I can sink for comfort while eating the beans on toast which was all I had the energy to cook for myself. One of the features of these cookbooks of mine is that they have no food stains on them. I don’t cook from them. When I have tried I’ve found that they have ingredients too exotic to be found in my kitchen or use pieces of equipment that I don’t posses. Like Methods papers they tell you about new and exotic approaches. Inspirational and aspirational, but if you try to emulate them you are going to need to work out a way to adapt them to the resources you have available in your lab/kitchen.
At the end of my shelf of cookery books there are three nondescript volumes whose covers are somewhat torn and whose pages are stained. These are the books I actually use when I want to cook something. Two are ancient volumes one of which has a twin printed in similarly garish colours, sitting in my mother’s kitchen, where it has been used constantly since the nineteen fifties, while the second has no pictures and I have no idea who the author is, if indeed their name is printed anywhere on it. The third, and most precious, isn’t a published text at all but is a scrap book assembled over the years from magazine clippings , give-away leaflets, supermarket promotions, and a host of jotted down recipes acquired from friends and family. These are the equivalents of Protocols. The recipes that are actually used every day in a researchers kitchen, sorry, laboratory.
At Nature Protocols we are trying to identify the experimental procedures that researchers will want to cut out and stick in their scrap books, metaphorically . We aren’t in the business of publishing the proofs of principle for shiny new experimental procedures. That is the remit of our colleagues on Nature Methods. No, at Nature Protocols we are looking for the techniques that experimentalist want to apply to their own investigations. These may be classic procedures tried and tested over many years, or approaches that were first reported a year or less ago. The most important thing is that they are procedures that will be used again and again.
We aim to supply the eager experimentalist with everything that he or she needs to perform identical experiments to those of our eminent authors. The Protocols are peer reviewed and edited to achieve a consistency of presentation which we hope make our protocols easy to apply in the lab. Also there is a substantial introduction to put the procedure in context, an anticipated results section to show how things are supposed to turn out and a troubleshooting table to help when things don’t go the way that one planned. Add to that a comprehensive list of reagents and equipment, and we think that our content is an invaluable resource for working scientists.
The commonest confusion is the difference between a method and a protocol. The simplest way to distinguish the two at least in my mind is this. If this is the first time that this procedure has seen the light of the scientific literature then you are not dealing with a Protocol, but with a Method. If you can point to one or more published papers in which the procedure under discussion has been used, then you are in the presence of a bona fide Protocol (and – which is more – you’ll be a Man my Son).