Remembrance of experiments past (a short guide to keeping a lab book)

Cornelis Bega – “The Alchemist”. (It’s a fair bet this dude is going to struggle to recapitulate whatever it is he’s doing. Also, his lab is a pigsty.) 

Keeping a lab book is a simple and utterly indispensable aspect of benchwork. TIR offers a few tips on how to make your record a reliable one.

It’s one of those things that seems obvious and intuitive, but is surprisingly easy to get wrong. A well-kept lab book can make all the difference to the working efficiency of a bench scientist, and can also serve as an important means of structuring your thought processes. Here are a few tips that may help streamline things.

What you need to note:
The essential feature of a lab book is that it should enable you to reconstruct and, if need be, exactly recapitulate any given experiment, no matter how long ago that experiment was carried out. A well-kept lab book cuts down on the probability of human error, facilitates troubleshooting, assists interpretation of unexpected results, and archives your thoughts and data. Some key things to record are:
What you did – what the process was (e.g. immunoblot, immunofluorescence, cell line generation, PCR, whatever). Note what you used (e.g. which cell lines, DNA, protein samples), and what the assay was if applicable.
When you did it – the date (this can be significant in terms of troubleshooting and in disputes – see below), and if relevant, the time (e.g. for inductions). Make sure there are page numbers so that you can index everything properly.
Why you did it – the aim of the experiment. What were you checking or testing?
How you did it – the full annotated protocol. This should either be written out in full, or, if it’s a routine procedure that you do at high frequency, note the alterations/customisations to a template protocol. Note absolutely every single detail, and every single deviation from the protocol, no matter how minor (e.g. extended incubation due to lunch) – these footnotes can be essential for interpreting an unexpected result, be it positive or negative. Never, ever trust your own memory! It is amazing how many details you can forget of a protocol once you stop doing it regularly. Don’t forget it won’t still be fresh in your mind months down the line.
Where you did it – details on the location of equipment etc can be important, especially if there’s a long gap between usage. It can be incredibly frustrating when you suddenly have to start trying to find a chemical or solution that’s sequestered in a refrigerator in an obscure corner of the department.
Where the data is now – most if not all of your raw data will be in digital form, so it’s important you note where it is to enable rapid retrieval if needed (e.g. in what folder the relevant image or Excel file is stored).
What your conclusion was – what did you conclude from the experiment? What decisions did you make about what to do next? (and if possible, note when that follow-up work was done)

How to produce the lab book:
Basically, you’ve got two main choices:
– (1) either write it up in real time, entering all details directly into the lab book. This has the advantage of being fast, but may be harder to follow and is more likely to be untidy.
– (2) you take your initial notes on scrap paper, and then write everything up neat afterwards. Although considerably more time-consuming, this has the advantage of forcing you to go over your data a second time after its acquisition, which can help put it into context and assists evaluation. One of the commonest traps for a bench scientist to fall into is to just generate data without sufficient analysis – writing everything up afterwards combats this to an extent. However, make sure you have all your initial notes dated and tidily collected in chronological order (either in a folder or inside the cover of your lab book – letting things pile up in a disorganised mess will not be helpful.

How to make it searchable:
A properly-maintained lab book is the hub of your personal research network – your raw data can be on a computer, the reagents in the lab, and so on, and the lab book links them all, as well as providing analysis. However…

– The big disadvantage of handwritten lab books compared to electronic lab notebooks (see below) is that it can be hard to retrieve information. When was it you carried out a particular experiment? Where is your analysis of a particular dataset? What is the usual dilution for a particular antibody?

– The best solution with a handwritten lab book is to have periodic indexes of your work. Classify each of your experiments/processes by category (e.g. cloning, biochemistry, microscopy/analysis, cell line X, etc), and note the pages. Then produce a single index page that groups all of the work done in each category over a given time period and what pages the entries can be found at. The frequency of index pages will depend on the size of the lab book your department provides, but every 70 pages or so is a good target.

What to avoid:
– Though time-consuming, a well-organised lab book will ultimately enhance the overall speed of your work by increasing efficiency. You will end up having to repeat things far less often if you know exactly what you did first time around.

– What you do NOT want is a situation where you are unsure what you did, or can’t retrieve the relevant data or information, and end up having to repeat the experiments in order to obtain it. Remember that the data for a manuscript can often be months or even years old when you come to assemble the figures, so it’s essential that you have well-organised and well-indexed entries in your lab book.

– Although the most important thing is that your lab book is fully intelligible to you, ideally it should be intelligible to everybody. Again, it’s entirely possible that your data will end up being used in a publication after you have left the research group you were working in, and it speeds things up if the record you left behind can readily be followed by others.


Something to remember:
– Don’t forget that your lab book is a legal document – this can become extremely important when patent or fraud cases are being considered. One of the biggest advantages to maintaining an old-school handwritten lab book is the potential for accurate and unambiguous accounting of time. It’s also a physical copy that can be readily retained and not lost somewhere in the depths of a server.

How to join the 21st century:
These days, there are of course many ways of producing electronic lab notebooks (ELNs), and doing things digitally has a number of advantages.

– Most if not all of your raw data will be digital in form. Digital data is much more easy to search, copy, and share, so having your entire lab book digitally is simply keeping up with the times – no more messing around with scissors and tape/paper glue!

–  From a group leader perspective, going digital means no longer accumulating a miniature Vatican library heaving with shelves and shelves of lab books from departed lab members.

– For an excellent consideration of the different options available, and practical advice on how to get an electronic lab book set up, see these two entries from quantixednotes to the future, the soft bulletin.

– Currently, probably the principle challenge with maintaining an electronic lab book is retaining a command overview of what you’re doing. Also (for now, at least) it can be more time-consuming to produce in the first instance – it’s currently still quicker to manually annotate an image, although it’s likely that this shortcoming will soon go away.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s