FAQS

1) How do I let the search engine know my mass accuracy window and my elemental formula constraints?

There are no parameters that need to be set. The mass accuracy window of IndexSearch varies with molecular weight. If the experimental mass of your protonated or deprotonated molecule is off by more than this value, then the search engine will not find the correct compound.

Over these mass ranges the following mass windows (in milliDaltons) are used:
  86 to   99:   1.0
100 to 199:   1.9
200 to 299:   2.8
300 to 399:   3.6
400 to 499:   4.0
500 to 599:   4.2
600 to 699:   4.0
700 to 799:   3.4
800 to 899:   2.2
900 to 990:   1.0

More accuracy is required to search at higher masses, but fortunately less than 2% of the compounds indexed are between masses 601 and 990. Within the mass window, the scoring system takes into account the mass error of the fragment ions and having more accurate data will therefore help to pinpoint the correct matching compound in the index. The window actually used (in ppm) will print out on your report.

No elemental formula constraints are needed other those imposed by the mass spectral data itself. The indexes contain compounds between 86 and 990 daltons composed of the elements C,H,N,O,S,P,F,Cl,and Br. As a result, the Available Chemicals Directory index has over 500000 commercially-available compounds and the larger PubChem index has 10335443 compounds.

2) Will the Rational Numbers search engines work for every compound?

No, the success of the program is compound dependent. Compounds that give at least one fragment ion will work, but the more fragment ions you have, the better the results. Paradoxically, having more heteroatoms in a molecule makes it easier to distinguish compounds, even though more heteroatoms in a molecular formula increases the number of possible compounds within the mass window of the experimentally determined molecular weight. Steroids with only one or two oxygen atoms are about the most difficult compounds to distinguish because of their combination of a high carbon/heteroatom ratio, large number of fragment ions, and the very large number of compounds in the indices with the same formula. Less polar compounds like these are still best handled with GCMS.

3) What do you mean by an index?

The two Rational Numbers search engines both search indexes. The indexes are much smaller in size than the molecular structure databases because they contain only the minimal information needed for comparing mass spectral data to chemical structures. For example, there is not sufficient information in the indexes to draw any of the structures. This means that the results are ID numbers and the user must have access to the MDL Available Chemicals Directory and/or PubChem to view the chemical structures.

Besides the speed advantage of using an index, this allows us to find salts that otherwise would be missed. For example, hydrochloride and hydrobromide salts are indexed at the molecular weight of their corresponding free base. Sodium and potassium salts are indexed at the mass of their conjugate acids.

Another advantage of indexing is that it allows you to find quaternary salts such as acetylcholine, provided that these substances are present in the databases as single cations or chloride or bromide salts. Quaternaries are indexed at their "apparent ESI mass".

4) Will an EI version be available later?

Right now MathSpec, Inc. is focusing on using accurate-mass CID-type data because of the lack of libraries for LCMS. These search engines should work for EI with some modifications. In the future we also plan to add silicon to the list of possible elements and "mathematically" silylate the indices.

5) Is the software suitable for data from all types of high-resolution mass spectrometers?

Yes. Each spectrum is captured in WordPad or Notepad and saved as a text file, using the vender's "Copy Spectrum List" utility. The spectra are then zipped into a single datapak using a compression utility such as 7-Zip.

No, the search engines require isotope ratio data on the protonated or deprotonated molecule because this data will restrict the elemental compositions that are possible. For an excellent discussion of the impact of isotope data and other important considerations, see The Seven Golden Rules.

On some instruments the centroiding parameters needed for optimizing accurate mass are not conducive to getting good isotope ratios; two different sets of parameters are really needed. For example, centroid 80% height with Np = 0.7 may improve mass accuracy, but for optimizing isotope ratios, try centroid 99% area and disable the time correction factor. Most Q-TOFs are capable of giving excellent isotope ratios when the proper parameters are used.

10) Has anything been published where I can get more details about the theory behind this search-engine approach?

Analytical Chemistry 2003, 75, 5362-5373.