Security News
Fluent Assertions Faces Backlash After Abandoning Open Source Licensing
Fluent Assertions is facing backlash after dropping the Apache license for a commercial model, leaving users blindsided and questioning contributor rights.
Read bruker data files using python with the speed of rust!
# pip install timsrust_pyo3
If what you want is to read spectra from a file, you can use the
read_all_spectra
function.
import timsrust_pyo3
datafile = "tests/data/230711_idleflow_400-1000mz_25mz_diaPasef_10sec.d"
specs = timsrust_pyo3.read_all_spectra(datafile)
This will return a list of Spectrum
objects. Which are the results of
“flattening” the frames into its individual isolation windows.
specs[0]
Spectrum(
index=0,
mz_values=[116.25158757494638, 118.9080472754818, 141.8956868151955, 167.55719727130426, 195.07915917591367, 232.07843727809686, 256.9100001557801, 267.30801795707947, 272.49521871359474, 274.26560752950985...len=1599],
intensities=[9, 9, 9, 9, 9, 9, 9, 9, 9, 9...len=1599],
precursor=Precursor(mz=812.5, rt=0.4161, im=1.1620169252468266, charge=None, intensity=None),
collision_energy=42.80258899676376, isolation_mz=812.5, isolation_width=25)
This shoudl work fine for a lot of cases.
If you want a more … raw representation of the data, we can use the
FrameReader
class.
reader = timsrust_pyo3.FrameReader(datafile)
all_frames = reader.read_all_frames()
print(all_frames[0])
print(all_frames[1])
Frame(
scan_offsets=[0, 0, 0, 0, 0, 0, 0, 0, 0, 0...len=710],
tof_indices=[7269, 226179, 238688, 283353, 302607, 313423, 320067, 325868, 333879, 334217...len=242412],
intensities=[20, 35, 20, 89, 45, 115, 57, 57, 113, 98...len=242412],
index=1, rt=0.33491, acquisition_type=DIAPASEF, ms_level=MS1, quadrupole_settings=QuadrupoleSettings(index=0, scan_starts=[], scan_ends=[], isolation_mz=[], isolation_width=[], collision_energy=[]), intensity_correction_factor=0.013324805457840315)
Frame(
scan_offsets=[0, 0, 0, 0, 0, 0, 0, 0, 0, 0...len=710],
tof_indices=[278272, 230453, 89438, 376113, 223326, 110881, 66872, 316096, 353458, 135560...len=4501],
intensities=[9, 9, 9, 9, 71, 9, 9, 9, 9, 9...len=4501],
index=2, rt=0.4161, acquisition_type=DIAPASEF, ms_level=MS2, quadrupole_settings=QuadrupoleSettings(index=1, scan_starts=[34, 370, 535], scan_ends=[370, 535, 708], isolation_mz=[812.5, 612.5, 412.5], isolation_width=[25, 25, 25], collision_energy=[42.80258899676376, 32.284789644012946, 25.174757281553397]), intensity_correction_factor=0.013324805457840315)
Note that here each frame does not have mz and ion mobility values.
lets start with the easy one … each tof index can be converted to a mz
value using the Metadata
class.
# We point the metadata to the analysis.tdf file
metadata = timsrust_pyo3.Metadata(datafile + "/analysis.tdf")
mzs = metadata.resolve_mzs(all_frames[0].tof_indices)
print(all_frames[0].tof_indices[:5])
print(" Becomes >>>> ")
print(mzs[:5])
[7269, 226179, 238688, 283353, 302607]
Becomes >>>>
[111.70270406113804, 767.4671439486657, 822.673531755788, 1035.4396805033862, 1134.6976413752595]
Now the harder one … ion mobility …
This is because each frame stores the mobility information by using the
scan_offsets
… which means … all peaks from scan_offsets[0]
to
scan_offsets[1]
are from the same scan, and thus have the same ion
mobility. and the 1/k0 value of that scan (0) can be converted as well
by using the Metadata
class.
scans = metadata.resolve_scans([1,2,3, 700])
print(" [1,2,3] Becomes >>>> ")
print(scans)
[1,2,3] Becomes >>>>
[1.3689703808180538, 1.3679407616361072, 1.366911142454161, 0.6492665726375176]
Note that since the tims funnel “elutes” values with larger 1/k0 first, the first scan is actually the one with highest 1/k0.
FAQs
Unknown package
We found that timsrust-pyo3 demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
Fluent Assertions is facing backlash after dropping the Apache license for a commercial model, leaving users blindsided and questioning contributor rights.
Research
Security News
Socket researchers uncover the risks of a malicious Python package targeting Discord developers.
Security News
The UK is proposing a bold ban on ransomware payments by public entities to disrupt cybercrime, protect critical services, and lead global cybersecurity efforts.