Jacknifing This is a method of resampling data in an effort to assess confidence in the hypothesised relationships between taxa.  It is very similar to bootstrapping. The original data matrix is sampled, with a portion of the data being used in order to create a new matrix.  In practice, usually 50% of the original data are chosen at random to make a new matrix.  Phylogenetic relationships are reconstructed between the taxa and the sampling process is repeated (many times, usually 100-1,000 times).  Finally a consensus tree (usually majority-rule) is constructed from the optimal trees from each pseudoreplicate.  The number of times a particular internal branch is observed is an indication of how much confidence we can place in the inferred relationships.
 

Choose the first letter of your term.
 

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

Any Questions? email: James Mcinerney