Using LIMS in an Academic Lab – Why and How?
All laboratories produce something, whether samples or spreadsheets. Ideally, every step in the process of making the product, including the origins of precursors and the physical or digital locations of the products, should be documented. I have worked in seven different academic labs, and they all dealt with this documentation in different ways and with varying degrees of success.
In my experience, most labs tend to leave documentation up to the individual lab members. Although there might be official lab processes for this, there is little, if any, monitoring of compliance. When someone leaves a lab, most of their samples will likely have to be chucked out because labels are illegible or ambiguous, sample identities are not cross-referenced to databases or notes, and paper notebooks may or may not be complete or legible. This is wasteful, because successors to the project might have to duplicate samples/reagents.
Furthermore, proper documentation of reagents is necessary for sharing with other labs. On numerous occasions I have requested plasmids, and they are rarely accompanied by a restriction map, never mind the sequence. Publications describing the reagents are not particularly helpful – they often don’t describe in enough detail how strains are made, and plasmid sequences or maps are even scarcer. Occasionally, I have to grow a new strain on various selective plates to determine its genotype. Plasmids are worse; it is usually up to me to clone the insert in silico, so that I can perform test digests of the plasmid and confirm its identity.
It’s unfortunate that most labs do not treat their reagents like a company product and their collaborators as customers. It always impresses me when labs send me reagents accompanied by thorough specification sheets. That level of professionalism is good PR. More importantly, good record-keeping protects the work from charges of scientific misconduct. When questions arise about the origins of a sample/reagent and how it was made, a digital record of the process is invaluable.
Laboratory Information Management Systems (LIMS) are designed to prevent documentation chaos. LIMS vary in functionality, but these software packages all aim to standardize sample/product information.
The LIMS system in the Peccoud Lab, LabCollector, is relatively generic and customizable, but there are many alternative LIMS, some of which are more specialized. I have been a researcher for over 20 years, but had never used a LIMS until I came to the Peccoud Lab less than a year ago. Since LabCollector (LC for short) is the first and only LIMS I have used, I will talk about how we use it in our lab. Much of what follows should be applicable to any biomedical lab.
LC is a web-based, password-protected LIMS run on the user’s server. There are several features of LC that we use extensively. First, LC has customizable modules for each type of entry. In our case, we have modules for Strains/Cell Lines, Plasmids, Primers, Sequences, Antibodies, Samples (intermediates) and Reagents and Supplies.
Each entry in LC is assigned a module-specific ID number. The information fields for each entry are also module-specific. For example, strains have names, genotypes, source, storage location and so on. Importantly, some of the fields are required to be populated before the entry is saved. You can also create cross-references to other entries via linked tabs, if, for example, one strain is derived from another. Some of the Modules (e.g. Primers) also cross-reference to Reagents and Supplies for reordering. Entries in the Sequences module can be pasted in or imported from Genbank, they can be annotated, and there are built-in BLAST and restriction map tools.
Second, LC allows us to print barcoded labels for each entry. This is handy not just for ensuring legible, freezer- and alcohol-proof labelling of tubes, bottles and plates, but it also allows one to identify a tube by scanning it. The sample name on the labels is usually sufficient for identification (more room than on a cap) and we rarely need to scan labels for most of the work we do.*
Third, every entry in the various modules can (should) be assigned to a physical location in the lab. The locations are customizable and can be anything from a cupboard to a specific position in a freezer box, with the box also having a position in a rack in a freezer. Freezer entries are associated with a graphical storage map, allowing one to determine the identity of a sample from its physical location in a box as well as its label.
Fourth, we use the LIMS to keep track of ordering information. Entries under the Reagents and Supplies module include supplier, catalogue number, price, and so on. All ordered items are kept in a list and orders can be checked off as they arrive. This feature saves a lot of time in finding ordering info, it prevents duplicate orders, and it ensures continued availability and consistency of supplies.
Finally, LC has mass upload features that allows one to import multiple entries or storage locations by importing a spreadsheet. This saves a lot of time on big projects.
LC also has various plug-ins such as an electronic lab notebook (ELN). We don’t use the LC ELN in our lab, I think because we started using a wiki system for lab notebooks before this plug-in was available. Having a built-in notebook system would be preferable though, as it makes it much easier to cross-reference lab notebook entries to LC entries. Of course, the ELN has to be flexible enough to handle the import of many different file formats for various types of data.
I have only mentioned LIMS features that any biomedical “wet” lab would implement, but most LIMS have additional functions that are tailored to specific clients such as core service labs.
There are a couple of methods of assessing compliance with the LIMS. A visual inspection of samples to determine if they have barcode labels can quickly confirm that they are entered in the LIMS. A second indirect method of assessing compliance is to perform a monthly audit of LIMS entries to assess if there are any unusual drops in the number of entries. There are any number of reasons for such a drop, but it could be because someone has not been properly documenting their work.
I stated earlier that LC is the first LIMS system I have used, but it is certainly not the only system of sample documentation I have used. Although rare now, some labs still use paper notebooks to document samples. This is the worst possible documentation method (other than none), for several reasons. Paper records are subject to physical damage, are more work to maintain (no ability to copy and modify similar or derived entries), and it is difficult to cross reference entries. Most labs simply use spreadsheets, sometimes with each person keeping their own (not good for uniformity of information), sometimes as a lab database on a server (better for uniformity). Spreadsheets allow for easy copying of information over to related or derived samples, cross-referencing is possible (although not easy), and if the spreadsheet is a lab database there is no duplication and sample information is theoretically uniform. However, there is no way to ensure that all of the necessary information is included for each entry. A LIMS system ensures that critical information must be included in order to save the entry. Also, as I have mentioned, LIMS systems have many functions that spreadsheets do not have, such as built-in tools (e.g. BLAST), graphics for storage and plasmid maps, and generation of barcode labels.
Like the 5S Management System I mentioned in a previous blog, a LIMS system is not a necessary component in academic labs. Many labs perform well with spreadsheet databases and save the extra expense of a LIMS. However, implementation of a LIMS does help reduce duplication of effort and waste, and does boost lab productivity and confidence in the lab products.
*Barcode labels are especially helpful, though, if using sample tracking in workflows. LC is developing a Workflows Management plug-in that I presume streamlines task assignments, sample processing and tracking, and data entry and editing in LC.