Classifion 1.2 - User Guide and FAQ

Screenshots - More Details

Download Site 1       Download Site 2       Download Site 3       Buy Now $145.00

Where does it sit...

Almost every mass-spectrometry lab is equipped with spectral-library search program with tens of thousands of spectra of pure compounds. The problems begin when you have very complex substances or using mass-spec methods different from those used in the spec-libraries. In some cases, you don't need to know what exactly is in the sample, but simply does it belong to a class of samples you already measured. In all these cases, you need to use classification software. But most of them are general purposed and require certain knowledge of the statistical methods used.

Here Classifion sits, offering you a classification with specialized for mass-spectrometry environment and automation allowing you to "avoid the depths of contemporary statistics", so you can concentrate on your immediate work.

How does it work...

Classifion needs to be "trained" using several measured spectra per substance. The "training" itself could be conducted automatically or manually (supervised). The aim of the training is to extract substance specific information based on statistical characteristics, but not on the interpretation of the mass-spectra. Having some knowledge on the mass-spectra structure could improve classification precision. Mathematically speaking the extraction is down to reducing the dimensionality of the spectra variable space (to each mass corresponds one dimension). PCA is well-known such as technique, which offers other advantages as ordering the principal component by "significance" (useful for noise reduction).

Classifion can train itself entirely on autopilot. Just import your data, run Autopilot, sit back and enjoy the view. The results will show you how successful the training was, so in case of problem you should re-examine your data or do supervised training.

The feature which makes Classifion better than the most of PCA software out there, is the use of proper factor space for each substance separately, instead of traditional approach of using common factor space for all substances. That improves the classification precision and makes the work with Classifion similar to the spectral-library search software.

The software...

Classifion is designed to work as stand-alone application as well as out-of-process COM-server. The latter allows Classifion to be used remotely from another application combining or not the remote access with user access.

Step-by-step example

The objective: based on their mass-spectra to classify an unknown sample to be (or not) one of number of known substances.

That example follows the shortest, most automated way for classification. That will work in the majority of the cases with decent data. Classifion will estimate the quality of your data for you.

1. You have to take at least 7 measurements per substance (assuming 3 or more substances). The typical number of measurements is 15 (+/-5). Keep the same conditions (as usually) for all the measurements.
2. Examine the data visually and dismiss the ones which are obviously bad. Including them in the training set probably won't harm the classification, but will increase the processing time for optimization.
3. Convert the mass-spectra into XY (mass/intensity) ASCII files with ".txt" extension. The eliminator between X and Y values is tab character (ASCII 9) or space character (ASCII 20). It is recommended that all the files to be in the same directory.
4. Open Classifion, and create new spec-tree (File menu).
5. Create separate group for each substance (File menu) and add the respective mass-spectra into each of them.
6. Press from All groups for boundaries in Options (right). Check, if all the options are unchecked, only "Normalize" should be checked. Check in PCAMD module Optimize page, the preset to be Medium correction.
7. Open Autopilot from Macro menu. Press Fly and enjoy the view.
8. Examine the Autopilot results
   
   
   
   The first column is the results from the optimization: number of excluded as inconsistent (bad) spectra and the size of the cluster in dispersion units - latter should be around 1. Second column is the cluster size from classification, it must be equal to that in the first column.
   
   Third column contains type I errors, or false negative. There will be merely always that type of error, because of many reasons - different matrix effects, different measurement conditions, misalignments of the instrument, human errors, etc. Having this type of error does not necessarily mean that the classification works poorly, it could be one of the reasons mentioned. Of course, supervised training could improve the results.
   
   Forth column contains type II errors or false positive. If you have that type of error, that usually means that training needs to be supervised.
   
   The final purpose of your classification will define which type of error is more important, hence optimized.
9. After you having your list of training sets in Classify module, you can classify any other sample you have. Just load and select the group with unknown specs (more than one measurement of your sample is recommended) and click Analyze Act. Group.

Screenshots - More Details