| Entering and manipulating data |
PAST has a spreadsheet-like user interface. Data are entered in an array
of cells, organized in rows (horizontally) and columns (vertically).
Entering data
To input data in a cell, click on the cell with the mouse and type in the data. This can only be done when the program is in the 'Edit mode'. To select edit mode, tick the box above the array. When edit mode is off, the array is locked and the data can not be changed. The cells can also be navigated using the arrow keys.
Any text can be entered in the cells, but almost all functions will expect numbers. Please note the decimal point convention which has been chosen by Windows depending upon your nationality: A comma (,) or a dot (full stop). 'Dotted' data can be 'commatized' from the Edit menu. Absence/presence data are coded as 0 or 1, respectively. Any other positive number will be interpreted as presence.
The convention in PAST is that items occupy rows, and variables columns. Three brachiopod individuals might therefore occupy rows 1, 2 and 3, with their lengths and widths in columns A and B. Cluster analysis will always cluster items, that is rows. For Q-mode analysis of associations, samples (sites) should therefore be entered in rows, while taxa (species) are in columns. For switching between Q-mode and R-mode, rows and columns can easily be interchanged using the Transpose operation.
Commatize
Converts all full stops ('.') in the data matrix to commas (','). This may be necessary for the program to read decimal points correctly, depending on your nationality.
Selecting areas
Most operations in PAST are only carried out on the area of the array which you have selected (marked). If you try to run a function which expects data, and no area has been selected, you will get an error message.
Renaming rows and columns
When PAST starts, rows are numbered from 1 to 99 and columns are labelled A to Z. For your own reference, and for proper labelling of graphs, you should give the rows and columns more descriptive but short names. Choose 'Rename columns' or 'Rename rows' in the Edit menu. You must have selected the whole array, or a smaller area as appropriate.
Increasing the size of the array
By default, PAST has 99 rows and 26 columns. If you should need more, you can add rows or columns by choosing 'More rows' or 'More columns' in the Edit menu. When loading large data files, rows and/or columns are added automatically as needed.
Moving a row or a column
A row or a column (including its label) can be moved simply by clicking on the label and dragging to the new position.
Cut, copy, paste
The cut, copy and paste functions are found in the Edit menu. Note that you can cut/copy data from the PAST spreadsheet and paste into other programs, for example Word and Excel. Likewise, data from other programs can be pasted into PAST.
Remember that local blocks of data (not all rows or columns) can only be marked when 'Edit mode' is off.
All modules giving graphic output has a 'Copy graphic' button. This will place the graphical image in the paste buffer for pasting into e.g. Word or Corel Draw.
Remove
The remove function (Edit menu) allows you to remove selected row(s) or column(s) from the spreadsheet. The removed area is not copied to the paste buffer.
Grouping (coloring) rows
Selected rows (data points) can be tagged with one of seven attractive colors using the 'Row color/symbol' option in the Edit menu. Each group is also associated with a symbol (dot, cross, square, diamond, plus, circle, triangle). This is useful for showing different groups of data in e.g. ternary and scatter plots and dendrograms.
Transpose
The Transpose function, in the Edit menu, will interchange rows and columns. This is used for switching between R mode and Q mode in cluster analysis, principal components analysis and seriation.
Loading and saving data
The 'Load' function is found in the File menu. PAST uses an ASCII file format, for easy importing from other programs (e.g. Word) and easy editing in a text editor. The format is as follows:
| . | columnlabel | columnlabel | columnlabel |
| rowlabel | data | data | data |
| rowlabel | data | data | data |
| rowlabel | data | data | data |
Empty cells (like the top left cell) are coded with a full stop (.). Cells are separated by white space, which means that you must never use spaces in row or column labels. 'Oxford Clay' is thus an illegal column label which would confuse the program.
If any rows have been assigned a color other than black, the row labels in the file will start with an underscore, a number from 0 to 6 identifying the color (symbol), and another underscore.
The 'Insert from file' function is useful for concatenating data sets. The loaded file will be inserted into your existing spreadsheet at the selected position (upper left). Other data sets can thus be inserted both to the right of and below your existing data.
Reading and writing Nexus files
The Nexus file format is used by many cladistics programs. PAST can read and write the Data (character matrix) block of the Nexus format. Interleaved data are not supported. Also, if you have performed a parsimony analysis and the 'Parsimony analysis' window is open, all shortest trees will be written to the Nexus file for further processing in other programs (e.g. MacClade or Paup).