Peak detection trouble

[ShusakuASANO]

Hi, I'm trying to detect peaks in my GPC (gel permeation chromatography) data. However, no peak was detected in HappyTools. Does anyone know the reason or the proper setting? Screenshot and data are attached. Thanks!

GPC chromatogram.txt

[Tarskin]

Is the intent to separate the overlapping 3 (4?) peaks in the 10-25 minute range?

[ShusakuASANO]

Yes, probabry there are 3 peaks. At least, I want to separate into 2 around 18 min and 22 min.

[Tarskin]

This is an interesting measurement, as the implemented algorithm is a bit too greedy in determining breakpoints in the f'(x) function that is applied through the top of the major peak, stating that the time-points belonging to a single f(x) range from 18.558 to 18.582. I expect that this is because the data in those maxima is not as smooth as I would expect, I am still thinking on how to avoid the program to crash (which is what happens under the hood) but there is a relatively 'simple' workaround which involves smoothing the chromatogram twice and then performing the peak detection (with your parameters, except I left number of datapoints at 100), yielding the following annotation that indicates 3 majpr peaks in that region (green, red and purple) and several smaller peaks of which brown seems plausible (as purple is trailing differently from the rest).

[ShusakuASANO]

Thank you for your kind reply and suggestion. I could reproduce the peak detection as you posted, however, the peaks are different in pdf report as the attached picture. I did "Peak Detection" then "Save Annotation", and conducted "Quantify chromatogram" using the saved annotation.

Is the image correct as the quantified chromatogram? I'm confused with the gap between beautiful detected peaks and the pdf reports.

Thank you,

[Tarskin]

The PDF report's first slide shows the area under the trace; determined using the default quantitation method (which yields a more robust quantitation for low abundant analytes):

Alternatively, the method to get the quantified areas as shown in the peak detection method would be to rely on the 'Gaussian Intensities' that can also be determined (but is currently not shown in the 'overview' slide of the PDF report).

[Tarskin]

There is a possibility to include both in the PDF, one corresponding to either method of quantitation, would this be useful for you?

[ShusakuASANO]

Thank you for the information. I checked pdf and results file for the "Gaussian Area" and "Peak Area". Peak #5, which was clearly detected in "Peak Detection" resulted in negative value in "Peak Area" and zero value in the "Gaussian Area".

I think that there is a mismatch between my chromatogram and the HappyTools algorithm. Could you check the attached data and give some comments?

Thanks!

GPC chromatogram.pdf

Peak Area (Background Subtracted) 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 296.5999390638495 1037.361396418066 2489.855875395111 1104.2913947821303 -1166.3445486992275 -2102.8603125536215 157.58604834359957 143.31796535396757 48.788284520311144 215.7270414207622

Gaussian Area (Background Subtracted) 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram -3.858025010572419e-13 -7.349676423018536e-14 2138.728752793353 3539.2318467793543 9.851675031313789e-12 -2165.6487152754307 -250.62455012694065 237.16611885617743 221.87670968164613 199.22317872768525

Peak Noise (Standard deviation of integration window) 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 116.69142901303182 449.2689069236874 684.0835902537236 560.3136929171383 237.46537890180485 410.2562589972835 67.34796029373877 20.467876648305133 7.7910189773050975 70.8324178596875

Background 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 69.29143155086108 204.6078296926224 648.8153753130615 1063.5816547636848 555.1238910129622 200.55487706573498 126.36553177077988 82.87964290816939 119.3849769825844 49.9414287028729

Noise 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 0.9863923702708274 2.362447971293854 8.747691686860202 4.349101054665511 5.890260375782651 0.9803251854658355 0.5195883055811534 0.19998673819305535 0.0185065588102879 1.7917957424317146

GPQ (Gaussian Peak Quality) 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 0.0 0.0 0.8678702341385529 1.0 0.0 1.0 0.0 1.0 1.0 1.0

FWHM 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 0.0 0.0 2.4701108496452866 1.6377049642796873 0.0 3.2639258958838973 0.0 2.0372714290789085 2.0052400127955243 0.9823947405255229

Signal-to-Noise 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 431.4867066877723 667.4581488495845 230.62658098210596 393.9710387097246 149.72908580431823 1233.6927469889981 465.68782172509805 325.8054094411633 1436.4920114583904 130.22184800558884

Retention Time [min.] 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 0.0 0.0 18.40833 18.58333 0.0 21.4 0.0 30.03333 33.75833 37.35833

Retention Time Residual [min.] 1 2 3 4 5 6 7 8 9 10 RT 15.077358 16.225707 17.370213 18.584032 21.263348 22.92186 24.547974 30.031305 33.762033 37.355372 GPC chromatogram 15.077358 16.225707 1.0381169999999997 0.0007020000000004245 21.263348 1.5218600000000002 24.547974 0.0020249999999997215 0.003703000000001566 0.002957999999999572