At its core, a LIMS in and of itself is little more than a software system with a high degree of variability. The system is built around the precept that it can and should be flexible enough to emulate the various workflows that the particular laboratory plans to encounter. What we see when we look at various existing open source LIMS is that they are for the most part purpose-built systems. This means that they are built with a particular set of laboratory workflows in mind. Systems like these will fare poorly or badly when confronted with workflows that differ significantly from their own main design. Consider LIMS designed for biological sciences. The systems built here are similar to more generic, commercial systems but are geared toward activities researchers and laboratory professionals in the biological sciences often utilize, like a BLAST search link or button on various screens. You will likely not find this in a commercial LIMS.
With a system like TWiki we have the opportunity to reshape the LIMS entirely. Rather than a single tool which has been crafted to solve a particular set of problems we have an extensible toolbox and a few instruction manuals. If you are not inclined to build things with basic parts then this type of solution is certainly not going to be for you at this time.
Before continuing let me explain that I'm going to target computerizing an assay that you find in education. My guess is that laboratories in this area will be more interested in trying to utilize a wiki as a LIMS. Secondly, I'm actually forced to limit the scope to something simple because LIMS is such a large subject that an attempt to cover something like the FDA's E. coli assays for bottled water simply will not fit inside of a small article.
Dartmouth College was kind enough to supply an assay for the identification of an unknown liquid by its density online, so we'll use it. There are probably (likely) dozens of ways to do this with more advanced instrumentation with far greater accuracy but this one will suffice for our purposes here. This test is good because it has multiple steps, a setup step, and a defined calculation at the end.
A wiki itself is a collaborative tool meant to allow diverse sets of people to easily communicate and present data. This data is normally organized in paragraphs. For a LIMS the TWiki wiki offers the advantage of both structured and unstructured data storage. The structured organization is what we are most accustomed to when discussing a LIMS. Most systems today use a relational database management system on the back end irrespective of what the data to be stored actually might be. In fact, a great deal of information needed by the laboratory and those who interact with its data is, in fact, hierarchical in nature and does not fit well into the RDBMS paradigm. Things like notes --instructions stored by laboratory personnel about the assay itself -- are all part and parcel of what individuals in the LIMS user communities have been asking developers to add to their RDBMS bound systems for years. A wiki (or something like it) is what these individuals wanted, but they did not have the requisite knowledge and information about such a utility to ask for one.
So with our Dartmouth unknown liquid identiy project we have a single sample of some unknown liquid. According to the test procedure we are to note the color and odor of the liquid and the room temperature first. Then students are to pour out the material into a clean, dry flask up to 10mL, weigh it, and then subtract the empty cylinder weight from the full weight. The difference is the mass of the liquid.
From a LIMS perspective there are several things needed to accomplish our job. For this exercise we will assume the following: 1) Individual samples of the unknown liquid will be given to a single student or a set of students working in a team setting 2) the instructions (test procedure) will need to be followed accurately in order to produce the correct results 3) according to the procedure the test must be run more than once and the results averaged.
The fact that the main samples of the unknown liquid will be distributed to students tells you that they can be uniquely identified. That is, Unknown Liquid A given to student team 01 can be different from the unknown liquid given to student team 02. If we give the samples themselves unique identifiers, let's say a number, we can easily differentiate them. You might be tempted to think in terms of simply labeling the various samples with letter names like 'A'..'Z' but numeric samples make more sense because samples given to student groups from different classes can be differentiated.
Thus far the requirements do not look very different from what is needed in a standard LIMS. Numeric sample identifiers, repeated tests; averaging of values -- this all looks and sounds like structured data. Where can we benefit from using the wiki features? Two areas immediately come to mind: the test procedure itself and the table of values the students will compare their results against. In the case of the latter, students themselves can improve the list of known substances by running the experiment repeatedly or by transferring the values from known sources. In the case of the fore, different types of students can add or reduce the test procedure information to suit them (think chemistry undergrads versus graduate students for instance).
In the next post of this series we will create the sample topics, associate them with forms to capture some pertinent class and student data, and then add the tests. It will become apparent how closely the wiki maps to the old paper test method's instructions, thereby facilitating the transition from paper to a LIMS.Go Back
Citation: Day Two Using Wiki as a LIMS Platform. (2011). Retrieved Mon May 1 02:11:53 2017, from http://www.limsexpert.com/cgi-bin/bixchange/bixchange.cgi?pom=limsexpert3;iid=readMore;go=1309635593