http://meddl.cs.wright.edu/ 2013-06-18T13:11:42-04:00 http://meddl.cs.wright.edu/ http://meddl.cs.wright.edu/lib/tpl/vector/images/favicon.ico text/html 2013-03-28T18:58:45-04:00 highlander http://meddl.cs.wright.edu/doku.php?id=config&rev=1364511525&do=diff MeDDL is a highly configurable tool. You tailor it to your specific equipment with the peaks.ini file. Furthermore, you configure MeDDL to your particular study by specifying the characteristics of your data with the experiment.ini file. Within each characteristic, the you then define the specific values that you wish to analyze. For example, a common data characteristic is time, wherein the you could describe samples taken at 0, 24, 48, and 72 hours. There is no limit on the number of character… text/html 2013-03-28T18:53:48-04:00 highlander http://meddl.cs.wright.edu/doku.php?id=downloads&rev=1364511228&do=diff The minimum recommended system requirements for MeDDL are as follows: * 1.8 GHz Dual-Core Processor * 4 GB RAM * Resolution of at least 1280 x 800 [Matlab] MeDDL is offered as a compiled Matlab program. Download one of the versions of MeDDL below, run the executable, follow the installation prompts, and run the file named MeDDL.exe. If you do not have Matlab installed, then you will be prompted to install the Matlab Compiled Runtime (MCR). The MCR allows you to use MeDDL without using a … text/html 2013-01-23T10:00:04-04:00 highlander http://meddl.cs.wright.edu/doku.php?id=links&rev=1358953204&do=diff * WEKA data mining tool from The University of Waikato text/html 2012-11-14T09:50:01-04:00 highlander http://meddl.cs.wright.edu/doku.php?id=about&rev=1352904601&do=diff Curt Grigsby, U.S. Air Force Research Laboratory, Human Biosignatures Branch: claude.grigsby@wpafb.af.mil Dr. Mateen Rizki, Wright State University Department of Computer Science and Engineering Chair: mateen.rizki@wright.edu Dr. Mike Zmuda, Miami University: zmudam@muohio.edu text/html 2012-09-17T05:22:27-04:00 highlander http://meddl.cs.wright.edu/doku.php?id=documentation&rev=1347873747&do=diff Background The MeDDL platform is a suite of tools currently implemented in MATLAB v2010a (The MathWorks Inc., Natick, MA) which allows for registration of “peaks”, which are defined here as a single ion or measured mass/charge (m/z) at a given retention time, peak mass and chromatographic time alignment, and a suite of statistical tools selected for biomarker screening studies. In brief, the MeDDL tool reads in lists of CDF (common data format) conversions of the raw GC/MS data files, registers… text/html 2012-02-27T15:24:26-04:00 boone http://meddl.cs.wright.edu/doku.php?id=extern&rev=1330374266&do=diff [Import CSV] MeDDL supports loading of external data. This data must be in a .CSV file. From Figure 1 (that displays the Main Table), select the File menu, select Import Data .CSV, and browse to the files that you want to import. A new table will be created in Figure 1 next to the Main Table. The fields for the external data must match the fields of the loaded data exactly. From the new table, you can select rows and display either a Line Plot, Box Plot, or Data Table by clicking the respective… text/html 2012-02-24T14:30:54-04:00 highlander http://meddl.cs.wright.edu/doku.php?id=visualization&rev=1330111854&do=diff The Visualization Tab provides users with an interface that allows them to visualize the peak sets that they created using the Filtering Tab. Using the two drop down boxes, you need to select the partition that you used to create the peak set and the peak set itself. After this, simply click the button with the desired visualization. Choices include: text/html 2012-02-24T13:04:57-04:00 boone http://meddl.cs.wright.edu/doku.php?id=heat_map&rev=1330106697&do=diff [Heat Map] The Heat Map is one of the two standard figures in MeDDL. It gives users a graphical 2D representation of the overall mass-spectra for a given set of data. The x-axis is defined as the mass axis and the y-axis is defined as the time axis. The color of a point on the heat map indicates its relative intensity. As we are currently using the Jet Colormap, the less intense peaks are blue, the moderately intense are yellow-orange, and the most intense are red. The Heat Map is also an intera… text/html 2012-02-24T13:04:44-04:00 boone http://meddl.cs.wright.edu/doku.php?id=peak_data&rev=1330106684&do=diff The Peak Data tab contains the main data table and will also serve as the location of any external data files that you load. For more information about loading external data, see the entry on External Data. Main Buffer This tab contains the main table that displays a summary of all of the registered peaks in a given data set. This table allows you to preform a variety of actions on any number of selected peaks. At the bottom of the table you will see a row of buttons that you will use to furth… text/html 2012-02-24T12:59:20-04:00 boone http://meddl.cs.wright.edu/doku.php?id=error_filter&rev=1330106360&do=diff [Error Filter] The Error Filter allows you to normalize your data, remove isotopes, and outliers. Note that some of the filters on this page replace the data with NaNs. The PCA plots cannot be generated using data that contains NaNs. Outlier Filter Summary Removes peaks that are outside of the range of the n number of standard deviations from the mean. This filter inhibits the use of the PCA functionality. Users specify the number of standard deviations to accept. Data outside of this range w… text/html 2012-02-24T12:57:43-04:00 boone http://meddl.cs.wright.edu/doku.php?id=data_filter&rev=1330106263&do=diff The Data Filter tab contains all of the filtering methods that allow you to down select your data so that you can focus on significant data. These filters create objects called peak sets. A peak set can be viewed in the Visualization Tab or classified in the Machine Learning Tab. text/html 2012-02-16T23:24:27-04:00 boone http://meddl.cs.wright.edu/doku.php?id=blind_data&rev=1329452667&do=diff Blind Data MeDDL supports loading blind data. This blind data can be classified using machine learning techniques. Currently, these machine learning techniques include WEKA and a novel hybrid genetic algorithm. Loading the Data Select the File menu from Figure 1. Click first on the External Data menu and then select Register New .CDFs. You will be prompted to select the .CDF files that wish to register as blinds. After this enter the fields that you want to associate with the blind group in … text/html 2012-02-16T23:10:13-04:00 boone http://meddl.cs.wright.edu/doku.php?id=machine_learning&rev=1329451813&do=diff MeDDL offers users with several types of classification methods. These methods use Matlab and Waikato Environment for Knowledge Analysis classifiers. Each of these allow users to classify data that is internal or external to MeDDL. The internal data classification allows users to classify the peak sets that they have created using the tool. The external data classification is currently designed to process a comma separated value file (.CSV). All classification methods support classifying intensi… text/html 2012-02-14T15:24:46-04:00 boone http://meddl.cs.wright.edu/doku.php?id=screenshots&rev=1329251086&do=diff [Main] [Sample PCA] [Hierarchical Cluster] [Sample Spectra] [Box Plot] [Data Table] text/html 2012-02-14T15:22:26-04:00 boone http://meddl.cs.wright.edu/doku.php?id=overview&rev=1329250946&do=diff The following describes the necessary steps for starting the MeDDL program. Many of these topics are further described in the following pages. Please see them for a more in-depth explanation. For an interactive guide, please see the downloads page for a test data set. The functionality of MeDDL will be demonstrated throughout the user guide using this sample data set. text/html 2012-02-08T23:21:17-04:00 boone http://meddl.cs.wright.edu/doku.php?id=guide&rev=1328761277&do=diff User Guide Index Please select a link below or visit the forum. Getting Started * Quick Start * Configuration Files * experiment.ini * peak.ini * plot.ini * Session Window * Directories * Session Buttons General * Main Tabs * Heat Map * Peak Data * Error Filter * Display Filter * Group Filter * Data Filter * Machine Learning * Visualization * Adduct Filter text/html 2012-02-08T22:46:18-04:00 boone http://meddl.cs.wright.edu/doku.php?id=display_filter&rev=1328759178&do=diff The Display Filter tab contains an array of check boxes that are generated using the configuration files. Initially all of these check boxes are checked. This means that you will see the data from all of the appropriate files when you display a plot using one of the buttons located under the main table in the Peak Data table. If, however, you wish to only view certain files in these plots, you can uncheck the respective check boxes. The Display Filter tab also contains the settings that are used… text/html 2012-02-08T17:22:21-04:00 boone http://meddl.cs.wright.edu/doku.php?id=session&rev=1328739741&do=diff [Session Window] Upon start up the Session window will appear. This window contains fields for the directories that you wish to use. You may use the Browse button beside each text box to browse to the desired directory, or type the path directly in each text box. Once you configure your directories, you are encouraged to save them by clicking the 'Save Session' button. After a session is saved the directories will be loaded automatically the next time that MeDDL is run.