NTsys

NTsys is a program to calculate various comparisions between taxonomic data, e.g. microsatellite DNA analysis (MSDA) data.

Simple use of the program

1. Set up the input data file:
   For MSDA data, convert the intensities to 0=missing and 1=present for each band.
   Delete all irrelevant information in the file, but keep the first row with band lengths and one column with sample lables (or better one with species and the second with sample lables, which then are merged).
   Save the file as a text file (csv or txt). Use space as the field limiter.
   Each field must be separated by at least one blank space (not tab or comma).
   With the note pad, add a initial title line, starting with " as the first character.
   The second line should contain the type and size of the file with four numbers:
   1: Type: 1=rectangular matrix
   2: #rows
   3: #columns
   4: 0=no missing data present or 1=missing data present, followed by the number for missing data, e.g. 99 or 999
   2 and 3 can be followed by a letter L, B or E, indicating that a lable is added as a separate field, before, or after the entry. Easiest is to use B for rows and L for columns. See an example below.
   When going from Mac to PC, you have to insert a line break before each line
   
   Note, that ideally, you should set 99 missing data when data is missing (i.e. not simply 0 everywhere).

2. Start ntsys

3. Go to Dis/similarity tab.
   Select Qulitative data
   Find the input file (you need to show all files, instead of only NTS files)
   Check by rows
   Select a coefficient. J is Jacard index; M is Manhattan
   Set the name of an output file
   Positive and negative codes can be kept
   Click on Compute to start the program
   This gives you a Jacard similarity matrix
   Close the report listing window.
   
   alternatively: use Internal data (sim int); Manhattan; by row; compute

4. Go to Transformation tab.
   Select Dcenter
   Use the results from the previous calculation as the input file
   Select an output file
   Keep Square distances checked.
   Click on Compute to start the program
   This is needed before ordination
   Close the report listing window.

5. Go to Ordination tab.
   Select Eigen
   Use the results from the previous calculation as the input file
   Select an output file
   All the default values are OK, but you could use more dimensions or check Show details.
   Click on Compute to start the program
   This performs the PCA.
   Early in the output, you can see the eigenvalues and how much of the variation that is explained by the different components.
   
   You get the 2D-plot by clicking in the NTSYSpc window, lower left corner on the first button.
   You get the 3D-plot by clicking in the NTSYSpc window, lower left corner on the second button.
   Click on Options - Plot Options - Points - Labels - Labels to see the labels

6. Cluster
   SAHN
   läst Jaccard file
   out UPGMA

7. Mega-X
   gör träd

Example of input file

"Rubus data exported 20/1-19
1 4B 5L 1 99
144 146 148 150 152 #This is the column lables
tv-tv96 0 1 1 1 1 #First column is row lables, first species than collect id
tv-tv97 0 1 1 1 1
tv-tv98 0 1 1 1 1
tv-tv99 0 99 99 99 99 #Missing data

If you get strange problems, try to restart the program and run again.

0593095A97A02FCC
s/n = MDMS4121I