Skip to 0 minutes and 11 secondsI am Rick Dunn, I am one of the lead trainers on this MOOC and this is our Q and A session, so you have sent in a whole range of different questions and comments We are going to talk about some of those questions and give some more information on that. Before we do there have been a lots of interesting comments which all the educators have enjoyed reading those, so thank you very much for being interactive during the course, we do appreciate that because we do understand some of the difficulties within the course. One person commented that they have enjoyed the course although it seemed to be designed for active metabolomics researchers.

Skip to 0 minutes and 50 secondsThat is true when we first developed the course that was the goal but we have been really delighted to see that people who are not directly applying metabolomics are also taking part and finishing the course as well. So we thank you very much for that and we do enjoy your comments about how we can improve the course as well. So please do keep those coming in so that we can make it even better for the general population.

Skip to 1 minute and 14 secondsBefore we start looking at the questions, one general comment we have had from a number of you is that the volume of data we collect and the amount of time that it takes to process that data and analyse that data, and that is true. It might take us 1, 2, 3 or weeks to collect and then it might take weeks or even months to be able to process and analyse that data and derive some biological knowledge from it and that is the same for all big data omics, whether we are looking at genomics, transcriptomics, proteomics or metabolomics, its exactly the same.

Skip to 1 minute and 49 secondsSo typically for example in the Phenome Centre here a project would typically take about 6 months to complete from sample preparation all the way through to biological interpretation. So lets get into some of the questions, there has been a whole range of questions and we have chosen some of these. The first couple are from Bev Burden thanks you very much for your questions. So Bev has asked I would like to know if an individual metabolic phenotype exists among humans.

Skip to 2 minutes and 17 secondsI read an article entitled Evidence of differences in metabolic phenotypes in humans which was in the Proceedings of National Academy of Sciences USA, in 2007 and Bev is wondering if this is now a reality Well we know from more and more research that has been performed here in Birmingham and across the globe as well, that each of us are different metabolically. Some of us are very similar metabolically and its very difficult just by measuring metabolites just to define differences between two individuals that are the same age, gender, ethnicity, work and live in the same environment etc. within there.

Skip to 2 minutes and 54 secondsBut we can clearly see differences that are related to age, gender, BMI and ethnicity across different populations on a global/ large scale population. And one project that was mentioned in the course was the Huserment project and this showed differences related to age, gender, BMI. We have a lot of data now looking at differences as to whether people respond or do not respond to treatments and the metabolic influence of that. So we can all be seen and I certainly view every individual as having a different metabolic phenotype within there. A second question from Bev was, so Bev is a registered nurse with a background in cardiovascular clinical outcomes management.

Skip to 3 minutes and 37 secondsShe is interested in learning more about how metabolic variables affect risk factors and individual patient response to medications. So some of the metabolic parameters that we have probably not talked about in the MOOC but things like your cholesterol your VLDL etc within there. They have an important role to play in health and disease, not necessarily how medication is absorbed into the body and metabolized in the body but more on metabolic health overall.

Skip to 4 minutes and 3 secondsThere are some for example, p450 enzymes which are more active or less active for example in individuals, there is a whole range of p450 enzymes within there These metabolise drugs or are involved in the metabolism of drugs and if they are more active or less active, that can influence how quickly a drug is absorbed into the body, how quickly it is metabolized and how quickly it is excreted, you have different pharmacokinetic curves in how you respond to that drug as well. Though we are all individual within that as well. So thanks Bev for your questions on that. Another question from Alexandra Iakab, so Alexandra is curious about mass spectral imaging experiments and whether you can re-use one frozen tissue.

Skip to 4 minutes and 52 secondsWell you can in a way. So normally you would thinly slice each tissue into very thin slices and you would place them onto a plate and then you would perform mass spectral imaging on that. You would not normally apply multiple different experiments on that unless you were doing some very specific experiments But normally you would take different sections and if you wanted to you could then reuse that tissue again and again, because you might get 20 or 30 sections from that tissue within there. Lexandra has also asked what happens if you change the ambient temperature of the tissue and the tissue suffers different cycles of freezing and un-freezing.

Skip to 5 minutes and 33 secondsWell ideally you would not be doing that because we know and we have shown in the MOOC that if tissues are basically frozen and then brought back up to room temperature you have the opportunity for metabolic activity to reoccur to start up again at room temperature and therefore change the metabolic composition of that tissue So in all of that sample preparation you will keep your sample cold. From Gamal Akabani I hope I have pronounced your name right there. He has asked a couple of questions about quality control and quality control samples.

Skip to 6 minutes and 8 secondsThe first one is are there different stages of the quality control process so can you have a quality control of the system versus the quality control of the system versus the samples. You can in a way there are some papers that have been published now that have indicated specific metabolites or changes in metabolites which are indicators of a high quality or poor quality sample in itself. So there are examples of where we can use those especially blood and if blood has not been processed properly into serum or plasma.

Skip to 6 minutes and 38 secondsTypically we apply a pooled QC sample and we use that to assess the quality control process or to apply the quality control process both both for the instrument and for the sample together and that is a normal process. We are using the quality control, we want to show, demonstrate the quality of the data across all of those processes. Gamal has also talked about different types of QC samples and the ideal QC sample is a pooled QC sample why a mixture of all samples? Well there is a very good reason for that.

Skip to 7 minutes and 16 secondsSo if we have got a hundred blood samples for example, plasma or serum samples and we want to be able to, that QC, that pooled QC sample, we want it to be as identical as we can to all of the biological samples that we are going to analyse. Both in the metabolites that are present in the pooled QC sample, their concentrations, we want that to be the median or the mean of all the concentrations across all of the biological samples and we want the sample matrix to be very similar or identical.

Skip to 7 minutes and 45 secondsthat is why we take all of those samples, we take a small aliquot of all of those samples and pull that together, rather than picking one or two samples and using a much larger volume of that for the QC sample. If we only take one or two samples then it is not representative of all of the biological samples that we will analyse. So that is why the pooled QC sample is out there. And keep an eye out in the literature there will be a publication published later this year that will talk much more about quality control samples and the quality control processes. The second question was are there different quality control standards for different systems and sample types.

Skip to 8 minutes and 22 secondsWell probably the best examples of where there is, so standard reference materials are reference materials which you can purchase for examples from NIST in the USA and they will be specific for example a specific sample type or can be specific to a certain disease as well. This is now being considered and discussed across the metabolomics community within there as well. So the most common one used at the moment is SRM1950 which is a plasma standard and there has been some papers published on that within there.

Skip to 8 minutes and 54 secondsSo there are different quality control standards, we do for example use some of those for long term assessment of our data over many weeks, months or years, and you can use it to assess data from different laboratories in inter laboratory comparisons as well. Another question from Gamal, can we carry out a real time metabolomics analysis of a live sample. Yes you can. There are a number of ways to be able to do that so if you are looking at humans you can for example look at breath and monitor that online in real time to define what is changing within the body.

Skip to 9 minutes and 32 secondsAnother example is the iKnife where for example in surgery the iKnife is basically measuring the plume of metabolites which are released as you perform the surgery. That is taken into a mass spectrometer and measured so you can actually measure changes in the tissue as you cut in real time and surgeons can use that to make sure they remove for example all of a cancer tumor within that operation. So there are a number and there are a number of other applications for example tracer analysis and flux analysis can also do that mainly in microbial systems. Another question, so do we use MS MS systems in metabolomics research.

Skip to 10 minutes and 14 secondsSo MS MS systems are tandem mass spectrometry so for example using triple quadrupole mass spectrometers and yes we do. So the targeted assays that we have talked about during the MOOC we would apply a triple quadrupole MS MS approach to be able to quantify metabolites in that system. We also use MS MS for gas phase fragmentation of metabolites to be able to identify metabolites as well. So MS MS has a number of really important applications within metabolomics. One final questions. So we have issues of accuracy and precision with mass spectrometry accuracy is related resolution, what about precision. So this question is really talking about mass accuracy and mass precision within there.

Skip to 11 minutes and 9 secondsNow mass accuracy, a general rule of thumb is the higher mass resolution the higher the mass accuracy that you can achieve within there. And the reason for that is the higher the mass resolution the narrower the mass peak is or the mass-to-charge ratio peak is and therefore the more accurately you can define the peak apex which is used to determine the mass-to-charge for that, and that is a rule of thumb within there. Mass accuracy is always dependent on the mass calibration that you perform, you can have high mass resolution but low mass accuracy if the mass calibration that you perform on the instrument is not appropriate and does not provide an accurate mass calibration within there.

Skip to 11 minutes and 49 secondsMass precision is not really talked about a lot, but typically on mass spectrometry once calibrated the precision of that mass ie, if you perform 20 different measurements of the mass the precision will be very high, the relative standard deviation for that mass error will be very low because mass spectrometers can operate in a very stable way to ensure that the mass calibration is stable within there. So one final comment rather than a question from Alexandra Iakab again. If we remove the peaks present in the blank sample could not we accidentally remove information that might be a possible metabolite with a same mass-to-charge ratio as the blank.

Skip to 12 minutes and 32 secondsVery unlikely, because if we are using, specially liquid chromatography mass spectrometry we are using both the accurate mass and the retention time and the possibility of there being a blank contaminant that has the same accurate mass and has the same retention time is very low. And how do we know whether is related to the sample or the blank, the whole reason for analyzing a blank sample is to ensure that we remove any features, any peaks, any signals that are related to or potentially related to a blank sample.

Skip to 13 minutes and 3 secondsSo in a one in a million chance in might be related to a metabolite but we have still removed that to ensure that we have got this high quality data that we can trust and if we make biological conclusions from that data we have high confidence in those. So thank you very much for registering and continuing onto the course.

Skip to 13 minutes and 25 secondsWe know some of you are still working through the steps, especially for week 4, continue doing that, continue asking questions online and we will continue to answer them and continue to interact with all the other learners as well within there So thank you very much for taking part in the course, we hope that you have enjoyed it and we hope to see you at a future course at the Birmingham Metabolomics Training Centre, thank you.

Course Q & A

Dr Warwick Dunn answers your questions from the course.

Share this video:

This video is from the free online course:

Metabolomics: Understanding Metabolism in the 21st Century

University of Birmingham