53


For volatile compounds, analysis via GC-MS produced highly fragmented mass spectra. These could then be matched to ‘fingerprint’ spectra, again via library and searching the NIST database. They found targeting a particular analyte was more reliable than totally relying on automatically-generated matches.


A further method to minimise Excel ‘clicking’ to reformat results was to use the open source software KNIME.


To finish preparing data, they would sort and tidy, ‘componentise’ then normalise to reduce amplitude variations from sample dilutions and spectrometer response. However, it was argued that this can reduce useful variation. Use of isotropic internal standards helped verify the validity of their data.


Compared to ca. 30 min for Excel ‘clicking’, KNIME gave workflow advantages: it was much quicker, taking just 2-3 seconds, and they found it was less error-prone.


Earll described a KNIME plugin called ‘mass cascade’, which is compatible with PCA. They used this to find soft ionisation [M+H] peaks for the flavonoid, naringenin.


Quality control of their data employed SIMCA PCA software, and their long-term QC sample was frozen tomato ketchup. They would inject a pooled sample over 10 dilution points, to encourage artifacts to elute, though it was sometimes difficult to distinguish between samples and impurities.


In collaboration with Nottingham University, they used a Thermo LTQ Orbitrap to study the ripening process of Ailsa Craig tomatoes. They used orthogonal partial least squares (OPLS) analysis, and found that metabolomes of the two tomato types diverged over time.


The conclusion from the talk was that the automation they implemented improved consistency of data and that MZ mine was very useful for unattended deconvolution of data.


Talk 6: MSMS prediction is it luck? Can a deeper understanding of 3D ionic structure help de-mystify this process?


A most illuminating talk was given by ex-Pfizer scientist Frank Pullen, from the University of Greenwich. He started with a brief reminisce back to mass spectrometry in the 1970’s, describing a number of rules which dictated the fragmentation pathways which resulted from the more-destructive ‘hard ionisation’ technique electronic ionisation (EI). Making the point that applied MS later began to demand applicability to more complex polar compounds, particularly pharmaceuticals, he moved swiftly on. He briefly went over modern soft ionisation ESI, where droplets are charged as they emerge from the electrospray needle, then like-charges accumulate at the droplet surface leading to coulombic explosions, formation of small gas-phase ions, typically [MH]+ for positive ESI. A brief overview of the detection process followed, using a triple quadropole MS as an example: the first ‘quad’ (as referred to by typical swaggering MS users) or Q1 tunes the stream, the second (Q2) enhances collisions, and is radio-frequency (RF) only. The third (Q3) then scans for product ions, leading to the mass / charge detector.


To explain product ion traces, there were key ions to look for in fragmentation patterns: carnitine, sulfonamide and amino acids.


Moving to prediction, it seemed that curly arrow theoretical mechanisms couldn’t explain loss of sulfonamide groups. Therefore his group looked to density function theory (DFT), based on quantum mechanics, for answers. Other prediction software was inadequate, predicting up to 20 product ions for a simple molecule.


Their approach was a fairly intuitive one, where DFT predicts which groups are likely to be protonated by soft ionisation. By theoretically protonating groups such as amines, the software calculates the bond elongation around the protonated group. The example molecule he showed had a tertiary amine group connected to a pair of isopropyl groups. DFT predicted protonation at this amine, causing loss of isopropyl groups which then resulted in a single fragment ion, as seen in practice.


For more complex species such as fluconazole, DFT predicted that the two most basic amines were most likely to be protonated. Thus amidazole ring was lost in fragmentation in ESI-MS. The next example was the Pfizer drug dofetilide, which contained three polar groups most likely to be protonated: a secondary amine between a sulfonyl and a phenyl ring, a tertiary amine and an ether linkage. The most basic secondary amine is protonated, leading to fragmentation at this point in the molecule.


Pullen concluded that DFT can clearly be used to predict MS-MS mechanisms, and that molecular 3D structure is part of the fragmentation. He said a quick DFT run can take as little as 30 seconds, leading to a potentially invaluable tool for the pharma industry to predict MS-MS.


Talk 7: Experiences in automating System Suitability


Adrian Dunn from GlaxoSmithKline in Stevenage discussed the benefits they have seen from implementing system suitability testing. At the core of this approach is a ‘test mix’ of compounds, to confirm system performance and give system-independent parameters. One benefit of this is a ‘value of merit’ when transferring methods between HPLC systems. This ensures an instrument is fit for purpose, to enable chemists to trust the data it produces and compare different instruments. Over time, trending plots enable problems to be flagged, possibly before a problem occurs, and resolve it promptly.


Historically, they used a four component test mix. Thus on a Monday morning, this was run on all instruments and the resulting data was analysed via spreadsheets.


With a focus on open- access chromatography with demands of high- throughput, introducing a newer automated approach made suitability testing less labour intensive: it reduced the time spent, and ultimately could be done remotely.


Approximately five years previously, their site moved to UHPLC and they changed to a seven


Gold sponsors, Thermo Scientific, were represented by Nicola Gardner, Senior Informatics Specialist


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60