SIRIUS command line with MS1 only data? about sirius HOT 7 CLOSED

With the latest release we introduce a command line parameter (--allow-ms1-only) that allows you to import data without MS/MS to run SIRIUS. It handles the TreeScore appropriately and uses only the IsotopeScore for ranking if no MS/MS is available.

from sirius.

I'm jumping in on this thread because I recently figured out this integration and thought I should share. I've been using R with SIRIUS for a while, but only recently updated to the latest version to get decomp. Now my formula decomposition workflow looks something like this:

mzs <- c(60.081324, 76.039854, 76.076239, 88.987485, 90.055504)

sirius_cmd <- paste0('sirius --noCite decomp',
                     ' --mass=', paste(mzs, collapse = ","),
                     ' --ppm=50',
                     ' --elements=CHNOPS')
sirius_output <- system(sirius_cmd, intern = TRUE)
sirius_output <- sirius_output[!grepl(pattern = "INFO", sirius_output)]
read.table(text = sirius_output, header = TRUE)

SIRIUS needs to be added to PATH for this to work, or you can specify the full path to the sirius.exe file. There's a bit of a startup cost to running it this way especially relative to Rdisop, so I'll usually pass it as many masses as possible at once.

Also @jorainer I've had some luck using SIRIUS with MS1 only by giving it the base MS1 peak as a single "MS2" scan. The .mgf files I write out from within R look like this:

BEGIN IONS
PEPMASS=60.0815010220745
MSLEVEL=1
CHARGE=1+
60.0815010220745 1247721.61759463
END IONS

BEGIN IONS
PEPMASS=60.0815010220745
MSLEVEL=2
CHARGE=1+
60.0815010220745 1247721.61759463
END IONS

and you can include additional isotope envelope things in the MS1 and leave the base peak as the only MS2 entry. SIRIUS doesn't seem to complain and I assume this is fine for the purposes of getting a chemical formula from a mass given an isotopic envelope, but someone here is welcome to correct me because I'm unfamiliar with the theory behind the magic.

from sirius.

Hey Johannes,
there is not much SIRIUS can do without MS/MS.

The only thing that would be possible is decomposing the precursor mass and scoring the resulting formula candidates
with the isotope pattern scoring.
All other methods rely on fragmentation trees, which need the MS/MS.

The mass decomposition is already available via the decomp standalone sub-command.
Isotope pattern scoring is currently not available for MS1 only data.

If this is something you would like to use from R we can make this available with the next build.

from sirius.

Thanks for the feedback!

IMHO it would be great to have this MS1-only formula prediction also available. Less for the real annotation processes, but I'm thinking of eventually improving the alignment and/or correspondence analysis in xcms based on that (or use it for quality assessment of the grouped chromatographic peaks across samples). We could also use Rdisop for that, but would be good to have an alternative - and it would be a first simple use case for the integration of SIRIUS tools into our R workflows.

Is the decomp already available or is this what you would make available next build?

from sirius.

That sounds great! Thanks for sharing this @wkumler (and sorry for my extremely late reply)!

Please also have a look at this @andreavicini

from sirius.

Nice workaround! Just be aware that the TreeScore which is part of the SiriusScore might be unreliable without MS/MS.
I would suggest to only use the IsotopeScore to rank the candidates.

from sirius.

Fixed with version 4.9.0

Feel free to reopen if this does not solve the issue.

from sirius.