Transcripts of the Attorney General's Initiative on DNA Laboratory Backlogs (AGID-LAB) Working Group
Monday, October 21, 2002
FORENSIC AUTOMATION THROUGH ROBOTICS
MR. MORRIS: This is what we're doing at automation. What is automation the way we perceive it and what are the benefits that we would like to obtain from automating our DNA evidence? The first thing that we've accepted is this principle that the automated process will not be the same as the manual process. If one wants to look at automation and you want to simply automate according to the manual process that you are implementing at that point in time, that might not necessarily provide you with the optimal solution. We firmly believe in the use of robotics. This is derived from the word robata, which was defined by Karl Capek in 1921. This is the way he looked at it was forced labor or drudgery.
I think the point of this is that in a process which lends itself toward automation the removal of people doing repetitive processes out from the final result up front enables one to move that human capital into the post-analysis part of the process, into the interpretation, evaluation, and providing of those results in court. So we try to remove human intervention as far as possible into this process.
What is robotics? Two definitions that we have worked on are reprogrammable, multifunctional manipulator designed to move materials, parts, tools, or specialized devices through various programmed motions for the performance of a variety of tasks and then an automatic device that performs functions normally ascribed to humans or a machine in the form of a human.
The way we see it for robotics is the application of instruments operating autonomously in a Carestian axis system in a voluminous 3-D space under control of preprogrammed inputs. Automation, the process of coordinating robotic action with no or minimal human intervention towards a predefined goal for a given process; that is, the important thing is the removing human intervention as far as is possible to get the reproducibility of results which are really not obtainable in a manual process.
The taking of samples lends itself to accuracy problems, reproducibility problems, which has an impact in the results that are obtained and also the cost of rerunning those samples. In South Africa we don't have large amounts of money available to push at the problem. We want to get the job done right the first time.
This is important. The robotic system allows the scientist to load more samples, assay more plates, and walk away from that process for a longer period of time. What is important here is the provision of auto tracking and the recording of every single step. Every sample, every subsample, at any point in time in that process this can be provided in court if so needed. One of the most significant benefits of liquid handling, which is fundamental to the process, is not so much as the speed and throughput, but the precision and the reproducibility of the assays which are being performed.
What I would like to do now is I'm going to hand it over to Johann, who is going to take us through a number of parts of this. First of all, the important thing about automation and the challenge maybe for crime scene samples even more so than with control samples is the vehicle in which the sample is collected. If the vehicle for sample collection isn't standardized, it provides a real problem in automating. We will go through what we have done and what we have put down for our reference samples and give you some idea of the integration of the various components of the system. Very often we get information from suppliers of instrumentation who will say this piece of equipment is fully automated and that doesn't fit into our definition of automation. So it provides a lot more challenges to us to get this process finalized and then some insight into what we are proposing and will be implementing in terms of the crime scene samples, which are an even greater problem both in South Africa and in countries throughout the world. Thank you.
MR. VAN NIEKERK: Thank you, ladies and gentlemen. I think when Karl Capek wrote the book on universal robots, he actually had a forensic science lab in mind because one of the premises of working in a forensic science lab is the fact that you do things over and over and over again and then you start and do it again.
So for that purpose we decided about two years ago that we needed to look at our processes and automate as much as possible for the event of running into backlogs, which everybody internationally does, and this is what we would like to talk you about this afternoon and just quickly show you what it is that we've done so far and bounce these ideas off of you in informal discussions afterwards, which we are obviously looking forward towards. So this is a purposeful meeting for us, and we really do appreciate the fact that we can be here.
As Commissioner Morris has said, the input vehicle for any automated system or any process we want to improve the efficiency of needs to be standardized according to our perception of things. We are in the fortunate position of being able to describe and design our own evidence collection kits in South Africa, and basically, given the small country and the remoteness of where we sit on the southern point of Africa, we are capable of having one specific data collection kit for the purpose of evidence collection. This kit really provides at this stage for us all the necessary equipment that is required for the collection of evidence.
Form follows function. That is one of the mottos for us, if you will. Whether we look at the design of an evidence collection kit, whether we look at the design of an automated system, at the end of the day both of those two entities must integrate and form a whole. For that purpose we decided methodology follows requirement. Our requirement is quite simply the following: We need to analyze more samples in less time. We need to be able to analyze more samples with less money available. We need to be able to analyze more samples so that we can establish an effective database so that we can really become effective in the investigation as well as the prevention of crime.
The contents of our evidence collection kits of which I've got a couple of examples here which I would like to show you later if time permits, we basically have divided our forensic DNA database into two parts. As with the CODIS system is the offenders index and the forensic index, we have a crime index, which equates to the forensic index, and then a reference index, which equates to the offenders index.
What we have done is we have standardized some Decon swabs in the kits, a couple of catch papers and combs for the collection of evidence and hair and some auxiliary equipment which just facilitates the whole process of evidence collection. Then the reference index component of our kits basically consists of the Marshal cassette.
On this point you will see the word Marshal coming up again and again. Obviously given the one L, it has no bearing on the Marshall University. Lewis Marshal was the starting point to name this process that we have designed. It basically comes from U.S. marshals. What does a U.S. marshal do? Take a person into custody, transports him or her to a place of safekeeping, looking after him the whole way.
This is basically what the Marshal cassette is doing for us. I'll show you a picture of what it is now. Then this cassette contains FTA paper, and we have designed our entire reference processing system around FTA paper. I don't think I need to tell you what FTA is all about. It's such a well-known commodity.
Our kits are bar coded. As Commissioner Morris has said, sample tracking is quite important to us. It's not quite important; it's essential. So all of our kits are prebar coded by the manufacturer, which just facilitates the tracking of samples still in the lab. Just to give you an indication of the contents of the kits, we have produced a little device very similar to something which is available from Switzerland. It's a little cardboard box for the collection and safe transportation of swabs. We have medical practitioners collect samples from survivors of sexual assault, and these are then placed into these cardboard containers which are prebar coded, as I've said. These are folded and closed down and sealed each with their own individual and unique identifier, individual bar code.
