LC Series Analysis - sample range not found

37 views
Skip to first unread message

Isabel Elliott

unread,
Sep 18, 2024, 8:59:24 AM9/18/24
to BioXTAS RAW
Hi,

I'm trying to process some SEC-SAXS data using Bioxtas RAW. After setting the buffer in the LC Series Analysis tab, I then try and use the 'Auto' button to select the sample frames but I get an error stating 'Sample range not found'. Is this likely to be a problem with my data quality?

Many thanks,
Isabel

See below for images showing the same elution profiles in ATSAS Chromixs and RAW:

Screenshot 2024-09-18 at 10.55.56.pngScreenshot 2024-09-18 at 10.56.32.png

Jesse Hopkins

unread,
Sep 18, 2024, 9:55:27 AM9/18/24
to bioxt...@googlegroups.com, isabelg...@gmail.com
Hi Isabel,

RAW's automatic sample region search algorithm is more restrictive than CHROMIXS's. That doesn't necessarily mean that either is better or worse, but that they each work better in different circumstances. If you want to see a few examples of this you can take a look at the most recent RAW paper, where I explore this a bit more.

In your case, the data look reasonable to me. I would manually select a region around the peak. When you try to send the region to the main plot, you'll get a warning telling you what RAW thinks might be wrong with the selection. My guess is that you have relatively low signal to noise, and so you probably have an outlier frame or two (like the small spike right near the peak) that is preventing RAW from finding a large enough region automatically. It's also possible there's a trend in the Rg or MW (the Rg looks uniform but noisy, but MW might show something), which RAW could report. You can then decide whether you want to proceed based on the potential issues it flags, or adjust your region to try to mitigate the issues.

Hope that helps. Happy to answer more questions.

All the best.

- Jesse

----
Jesse Hopkins, PhD
Deputy Director
BioCAT, Sector 18
Advanced Photon Source


--
You received this message because you are subscribed to the Google Groups "BioXTAS RAW" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bioxtas_raw...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bioxtas_raw/0c5283bc-2109-4411-b1a2-afb347ed38f2n%40googlegroups.com.

Isabel Elliott

unread,
Sep 18, 2024, 12:24:32 PM9/18/24
to BioXTAS RAW
Hi Jesse,

Thanks very much for the reply. I've had a go at selecting a peak manually for my datasets, but I get a lot of warnings suggesting that most of the frames are dissimilar and also that there are trends in MW/Rg (depending on which frames I select). Would this suggest that it's not possible to further analyse this data?

Thank you,
Isabel

See below another example from a different dataset:
Screenshot 2024-09-18 at 17.20.33.png

Jesse Hopkins

unread,
Sep 18, 2024, 6:07:47 PM9/18/24
to bioxt...@googlegroups.com
Hi Isabel,

Generally speaking you're probably fine if there's only a few reported outlier frames. In the case you attached, I would recommend zooming in on the Rg across the peak (and the MW) and see if there's a noticeable trend, the large stuff coming out just before the peak makes it hard to tell what's going on. If you do see a trend in either of those, then you should attempt EFA on the data to try to deconvolve whatever is in the peak. You might want to do that anyways, because the message says there are two significant singular values, which means two components eluting in the range you selected. I'd definitely recommend EFA in this case.

If you see frame warnings, you might try plotting the subtracted profiles in your selected region in the main plot and taking a look at how different they are (they should be the same within a scale factor). This might give you a better idea of what's going on here. If you see systematic drift (such as in the high q, based on your warnings) that might indicate a drifting baseline, accumulating damage, or other such things that you need to compensate for in some way.

It might be that the data isn't usable, but I think you can probably get something out of it. You will have to carry out some more investigation to do that though.

Hope that helps.

All the best.

- Jesse

----
Jesse Hopkins, PhD
Deputy Director
BioCAT, Sector 18
Advanced Photon Source

Isabel Elliott

unread,
Sep 19, 2024, 4:41:39 AM9/19/24
to BioXTAS RAW
Hi Jesse,

Thanks again, this is really helpful. Is there a tutorial to show how to plot the subtracted profiles in the sample selected region from the SEC elution?

Best wishes,
Isabel

Jesse Hopkins

unread,
Sep 19, 2024, 9:20:21 AM9/19/24
to bioxt...@googlegroups.com
Hi Isabel,

Steps 34-37 of this tutorial go through sending series frames to the main plot:

All the best.

- Jesse

----
Jesse Hopkins, PhD
Deputy Director
BioCAT, Sector 18
Advanced Photon Source

Reply all
Reply to author
Forward
0 new messages