Search Tolerances

Several parameters can impact precursor and fragment tolerances during the database search phase.

{
    "precursor_tol": {
        "da": [
            -500,
            100
        ]
    },
    "fragment_tol": {
        "ppm": [
            -10,
            10
        ]
    },
    "isotope_errors": [0, 3],
    "wide_window": false
}

Precursor Tolerance

This parameters specifices whether an absolute ("da") or relative ("ppm") tolerance is used for selecting candidate matches. Sage applies the "left" side of the tolerance to the experimental mass - this means that typical open-searches (i.e looking for a positive delta mass arising from a chemical modification) will want to use a large negative value here. Using the above configuration, an MS2 spectrum with a 2700 Da precursor mass would search for candidate peptides with a mass between 2200 and 2800 Da.

A standard "closed search" might use the following configuration:

{
    "precursor_tol": { "ppm": [-20, 20] }
}

Fragment Tolerance

Similar to above, you can select whether to use absolute or relative tolerances:

For high-res MS/MS:

{
    "fragment_tol": { "ppm": [-10, 10] }
}

Or for low-res MS/MS:

{
    "fragment_tol": { "da": [-0.4, 0.4] }
}

Precursor Charge Range

If precursor charge states are not annotated in the mzML, or if the wide_window setting is turned on, Sage will attempt to search the given precursor m/z at multiple charge states (by default, z=2, z=3, and z=4). This setting can be configured to override this behavior.

Default value:

{
  "precursor_charge": [2, 4]
}

Isotope Errors

This parameter essentially runs a multi-notch or mass offset search for each integer value in the (left .. right) range by multipyling the integer value by the mass difference of one C13 neutron. This can account for incorrect assignment or sequening of the monoisotopic peak.

Default value:

{
    "isotope_errors": [0, 0]
}

Typical values:

{
    "isotope_errors": [0, 3]
}

Wide-window mode

This parameter instructs Sage to dynamically change the precursor tolerance for each spectra based on the isolation window encoded in the mzML file. This is useful for searching wide-window acquisition (WWA) data, parallel-reaction monitoring (PRM), or data-independent acquisition (DIA) data. You probably also want to update the following settings as well if you are running a WWA/PRM/DIA search:

{
    "wide_window": true,
    "chimera": true,
    // Try running multiple searches with different `report_psms` values
    "report_psms": 5
}