Your Suggestions

Please post your questions and suggestions for topics here!


15 thoughts on “Your Suggestions

  1. I have a question about capillary electrophoresis in the well known Amanda Knox case.
    The argument is set out here :

    Very briefly, the question is if the differences in the two capillary electrophoresis runs on the SAME sample suggest that the ABIPRISM 3130 was contaminated from earlier runs ( I think they do – the differences seem too big to be accounted for by noise ).

    I’d be interested on your opinion on this, thanks ( I was referred here by Chris H. )

    1. Hi George,

      Great question. Thanks for sending the link to the kermit-analysis. Well, the discrepancies between the two run results are not consistent with the same amplified product being run twice even if on different instruments and on different days. The inversion in peak heights simply should not occur if the same thing is being run multiple times. The results actually look more like different amplification products. In my experience, contamination of DNA from previous runs has never been an issue with capillary electrophoresis. The design of the instruments provides excellent safeguards from this. Of course, anything can occur. The best environment for something like this to happen would be if prior to the knife blade sample being run, a sample with a very high level of Kercher’s DNA was run. It’s hard say what is going on here, but for many other reasons the data is essentially useless.

  2. Thanks for your response, at least I can now feel that I’m not being completely stupid here!

    Due to the intense interest in the case, and the existence of web sites which do not accept Amanda’a innocence, I’m interested in digging as deep as possible to find the best explanations for the forensics. I appreciate it’s hard to say what happened, the objective is to come up with plausible explanations, as well as just observing that the evidence is useless (which I certainly agree is correct).

    So what are the possible explanations? It seems it has to be either some kind of a labelling or identification error ( so the samples tested are different, and both happen to contain Meredith’s LCN DNA ), or contamination. The first seems unlikely ( correct me if I;m wrong), so If it’s contamination, how could that happen? I suppose it could be in the preparation for capillary electrophoresis ( where the dyes are added I guess ), or in the 3130 machine itself.

    Are there any gross errors that could occur? What if the water reservoir was empty? DOes the “Water Wash Wizard” run automatically, are there any circumstances under which it might not run? Is there any way for material from the waste vial to re-enter the polymer path? Or could it simply be that the capillarys were not well cleaned after earlier runs, or waste material somehow was badly routed and didn;t end up in the waste vial.

    Another thing I’m looking for is example runs that show what should happen if you run capillary electrophoresis twice. Are there any examples of this online?

    Sorry for asking so many questions, you should understand that I have never even seen one of these machines, let alone operated one, I’m approaching this as an interested amateur looking for answers.

    Best regards,

    1. My pleasure George. Unfortunately, the only way to truly dig deep into the forensics is to have the data. Discovery is key in providing meaningful review in DNA cases and it seems like full discovery may never have been provided in this case. All I can work off of is the electrophergrams posted online that are barely legible. This is why I have to be somewhat vague in my analysis of things. I highly doubt that there is reasonable concern that the instrument itself would cause contamination. Again, if someone is wildly out of protocol, then who knows what could happen? Let me dig in my files, I may be able to come up with some examples that I can provide of multiple injections of the same sample. One thing to keep in mind is that capillary electrophoresis is in fact more sensitive than any of the quantitation assays, this is the reason that negative controls need to be taken all the way through the process, even if they are negative at the quantitation step.

      Another issue is that if the same contaminant is present in both injections….it too should have the same appearance during capillary electrophoresis in terms of peak height ratios. Here is a list off the top of my head regarding where contamination could occur (assuming that the result from the knife blade (still meaningless!) is not a genuine portrayal of what was on that knife when collected and is a result of contamination:

      -If during collection and storage it came into contact with a source of Kercher’s DNA
      -If it was extracted along with items containing Kercher’s DNA and became contaminated in the extraction phase
      -If it was being prepared for quantitation or PCR along with samples containing Kercher’s DNA and it became contaminated
      -It sounds as if the extract was concentrated…contamination is a concern here, and I’m not sure that we know if the appropriate controls were used in parallel to the concentration
      -If during prep of samples for capillary electrophoresis, samples containing Kercher’s DNA contaminated the vial

      I wonder if a log of contamination instances for other cases was ever produced. In general, if a lab says they have never seen DNA contamination, it means they aren’t looking hard enough. It happens.

      -planting? if so, a poor job was done!

      I suppose if there were some sort of radical difference in instrument parameters between the two runs, weird things could happen, but there is no credible way to assess things like that without the instrument log, etc. If in fact the same amplified sample were run with the same instrument parameters and I obtained the results in this case, as an analyst I would look into things further and be skeptical of the result. Also, in my experience I don’t give data less than 50 RFU’s much study as I would typically not consider it to be reliable.

      The fact is that when you are troubleshooting issues in DNA testing, sometimes you really can’t get down to the exact cause. Sometimes you simply have to repeat the analysis if possible or not report the results if questionable data arises.

  3. After your comment “The results actually look more like different amplification products.” I did some more thinking, and the result is this page which attempts to analyse whether the knife blade results can be explained by the sample being contaminated with PCR product, and two amplifications being performed. It seems that this is plausible, and I estimate the initial number of copies for each marker. The problem with this is that it contradicts court documents which state that the amplification was only performed once.

    I have another question though. Suppose we have a sample contaminated with say 100 PCR fragments (from an earlier run), and this is subject to PCR amplification. Is it possible that an initial uneven distribution could be preserved by the amplification process, that is the amplified sample might not be well mixed? If so, this could explain the differences in the two electrophoresis runs. I suspect the answer is No, and this is a stupid question, but I’m asking anyway 🙂

    1. Hi George,

      I also seem to recall that there was some confusion regarding whether the extract had been concentrated. I wonder if it was perhaps amplified before and after concentration. Although the hope in concentrating an extract for PCR is that you will get larger peak heights in CE, that is not always the case (for example, you may also be concentrating an inhibitor). Its safe to say that low level DNA amplifications are going to be highly susceptible to stochastic effects in PCR and re-amplification of a small sample of previously amplified DNA would not likely preserve the peak height distributions.

      PCR product can potentially contaminate other PCR products being prep’d for CE. No amplification necessary since the PCR product will have numerous copies of DNA already tagged with fluorescent primers. Keeping pre-amp and post-amp work functions physically separate is a given in any good DNA laboratory. Typically, separate rooms and one-way work flow, etc. virtually ensure that PCR product will not contaminate DNA extracts being prep’d for PCR. However, not all labs are alike with regards to QC.

      Within a post-amp work lab, however you do have PCR products in proximity to tubes being prepared for CE.

  4. I have a new question, as follows: during DNA extraction, before PCR amplification, do the chromosomes remain paired, or are they separated?

    The relevance of this question is to see whether the crucial knife-blade results in the Amanda Knox case are compatible with the sample (supposed to contain a few complete copies of Meredith’s DNA) being divided into after extraction, and being amplified in two wells ( that this division occurred is I think certain – a document has come to light that shows hundreds of examples, and in each case Stefanoni the sample is processed in two wells – there would be no reason for the crucial sample to be processed differently ), or whether the only possible explanation is that the sample was PCR contaminated.

    1. Hi George,

      During DNA extraction the nucleus is broken apart and all of the chromosomes are unwound into a DNA soup. Pairs of chromosomes are floating randomly in the DNA soup. The only time you see nice maps of chromosomes with pairs next to each other is in a karyotype ( But more importantly to your point….if two portions of extracted DNA are independently amplified, you would expect to find some differences in peak height ratio and perhaps in detected peaks if the level of DNA is extremely low. This is due to so-called stochastic effects in the PCR process.

  5. Hmm.. having done some further reading, I think I had a mis-conception – it seems chromosomes are not normally paired, and this only happens during mitosis and meiosis. I’m revealing my almost non-existent knowledge of cell biology! Apologies.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s