The first MOOC of its kind, this course is an introduction to metabolomics principles and their applications in various fields of life sciences.
We will provide a summary of all steps in metabolomics research; from experimental design, sample preparation, analytical procedures, to data analysis. The course also provides case studies of various kinds of research samples to attract students that are not familiar with metabolomics, providing them enough explanation to utilize metabolomics technology for their respective research fields.
Several examples of metabolomics applications will be introduced throughout the lectures. These include examples within food science and technology, metabolic engineering, basic biology, introduction to imaging mass spectrometry, and application in medical science.
No previous knowledge on metabolomics is needed but we recommend that students have an undergraduate-level understanding of Biochemistry, Analytical Chemistry, and Biostatistics, and that they learn about basic principles of multivariable analysis prior to taking this course.
I was wondering if anyone knows a way to get good centroiding of Waters data that was recorded in profile mode? In MassLynx you can centroid a single spectrum and it looks like this as an example:
That seems reasonable by eye. I tried then using the centroiding in msconvert (Proteowizard). Here is the result:
What it has done is not finding the center of the peak but has chosen the top scan for the m/z. This is not very accurate and is in this case a difference of 15 ppm.
So my question is if someone knows a better way to centroid data that was already acquired? Msconvert can use vendor algorithms for centroiding for a lot of formats but apparently it is not available for waters data.
UPDATE: msconvert now supports vendor (AKA Waters) centroiding.
In my dataset I get features > 1000 with no problems. Are you sure your data contains data above 1000 Da? If you are sure your data has peaks > 1000 Da your best bet is to post a reproducible example with code and sample data.
I haven't tried to manipulate it like that so I don't know.
But your peakwidth is too narrow I think if you want all in one. You have a max of 10 sec. If you want to merge all you need several minutes I assume. At least I'd try that.
One way to achieve this, I think, is to set mzdiff to something positive. That should make everything connected integrate as one. Of course this might have unwanted effects in other places.
I think this is gonna be very difficult to achieve with so much overlap. What I would try is lowering the minimum peakwidth. If that doesn't work last resort is trying matchedfilter instead of centwave.
More information: We would like to invite you to Bio&Data, the first workshop of the newly established MOVISS – “Mountain Village Science Series” taking place in Vorau, Austria (Sep 20-23, 2017). MOVISS Bio&Data is different to the usual conferences. It is rather constructed as a small, problem-driven meeting, full of discussions and questions about how to deal with metabolomics data reasonably. In this way, we hope to constructively engage some of the greatest minds collaboratively in solving some of the challenges of the metabolomics and bioinformatics community.
Four sessions are planned, each devoted to a separate step of the metabolomics process; Design of Experiments, Analytical Analysis, Data Processing and Statistical Analysis in the biosciences will all be discussed including your data if you bring them for discussions.
We plan a summary of this discussion will be produced as a paper for publication to share within the wider metabolomics community. Finally, you can continue with the R Summer Schoolfrom September 25 – 27, 2017 in Vorau!
We are delighted to announce that Early bird registration is open for Metabolomics 2017 – the 13th Annual Conference of the Metabolomics Society. We look forward to seeing you in Brisbane! Please visit the new website for more information and take advantage of the early-bird pricing.
Website:Metabolomics 2017 Hosted by: The Metabolomics Society Where: Brisbane, Queensland, Australia When: June 26-29, 2017
Abstract submission information will be published within the next month. We look forward to seeing your latest findings!
The Society has obtained a group block of hotel rooms, visit the Hotel page of the website to view several different options. You are encouraged to book your hotel accommodation as soon as possible, since June is a very busy time inBrisbane.
Happy New Year! The Metabolomics 2017 Planning Team
I have not been aware of intf since I am normally using centwave that doesn't have this concept. I am using into because when I compared to intb the CAMERA grouping was better with into suggesting that was a more stable measure.
As for intf vs into I cannot answer but I guess it depends how good the fitted model is.
pic3 looks like shoulder peaks. But difficult to see when I only have that zoom level. What you need to do is take one of the largest peaks, look at the spectra, zoom in around the mass peak at low intensity. If you see a lot of small peaks (1-5%) around the real peak --> that is shoulder peaks and you need to filter them before an analysis.
0.0005 sounds too narrow to be to compare peak tables. 0.0025 seems more reasonable. The tools might choose the mass differently (at apex, mean/median across the peak, mean/median across samples).
16238 seems like a lot of features. Suggests to me something is up. Could be if there are many shoulder peaks, if you detect a lot of noise, or chop up peaks like mzmine did.
I am not familiar enough mzmine to tell you what to tweak but it makes sense if the other tools give you 1 or 2 peaks for that noisy peak.
