search

MannLabs / alphatims

An open-source Python package for efficient accession and visualization of Bruker TimsTOF raw data from the Mann Labs at the Max Planck Institute of Biochemistry.
https://doi.org/10.1016/j.mcpro.2021.100149
Apache License 2.0
81 stars 25 forks source link

[Question] precursor_indices definition #265

Open jspaezp opened 1 year ago

jspaezp commented 1 year ago

Hello!

I have been exploring the package and had a question on how the 'precursor_indices' is defined. When I filter based on the precursor index (1 in this case), get two distinct quad isolation.

This is contrary to what I thought would be the case (I was expecting each precursor to be only one iso window), so I wanted to know whether this is an expected behavior or how the precursor index is being defined.

Is there any indexing to access unique quad isolation windows in a sequential manner?

Appreciating your help! -Sebastian

Snippet showing how I am getting the results

# This file comes from the PRIDE repo for the alphatims paper
# curl ftp.pride.ebi.ac.uk/pride/data/archive/2022/02/PXD028735/LFQ_timsTOFPro_diaPASEF_Ecoli_01.d.zip --output ../data/ecoli_timsTOFPro_diaPASEF.d.zip
dia_data = TimsTOF("../data/LFQ_timsTOFPro_diaPASEF_Ecoli_01.d", mmap_detector_events=False)
inds = dia_data[{"precursor_indices": 1}, "raw"]
out = dia_data.convert_from_indices(inds,
        raw_indices_sorted=True,
        return_quad_mz_values=True,
        )
np.unique(out["quad_low_mz_values"], return_counts=True)
# (array([400., 800.]), array([ 20845610, 119548979]))
sander-willems-bruker commented 1 year ago

Dear @jspaezp . This is the original intended behavior (see page 18-19 of the GUI manual). I understand that this indeed can be a confusing situation, but at the time I felt it was the best option since "precursor_indices" is first-and-foremost a DDA feature (if only by using the name "precursor") and only an afterthought for DIA. After the initial implementation two years ago, I now agree that defining DIA precursor_indices as a single "window" might make more sense than the current definition of a "window group". That said, with new acquisitions such as synchroPASEF, the "window group" definition might again become the more dominant use case. As such, I will leave it unchanged for now to ensure backwards compatibility and avoid any further confusion.

With respect to your second question on how to select unique quad isolition windows in a sequential manner, I personally tend to use quad setting directly to do this. Notice that you should be able to find all those easily by viewing the analysis.tdf with any sql browser, or by checking out the .cycle attribute of a TimsTOF object.

jspaezp commented 1 year ago

@sander-willems-bruker Thank you very much for the reply! that makes a lot of sense now! I think it would be great to have some of this info in the python API documentation. I am not sure if I just missed it or it it is too small of a user-niche to invest in it.