How the IDH Trees Were Produced

Here is how the IDH phylogenetic trees were produced:

1) The original 24 IDH sequences were submitted to Clustalw for multiple alignment, with default parameter values and PHYLIP formatted output.

2) The first 150 positions and final 145 positions were truncated from the multiple alignment. Since the sequences were of different lengths, this focused the phylogenetic tree computation on the core of the alignment (550 positions).

3) PHYLIP Version 3.57c was used to infer the phylogeny of the 24 IDH sequences from the multiple alignment. The specific steps were:

a. Run SEQBOOT on the multiple alignment file, producing 10 bootstrap replicates (samples from the sequences).

b. Run PROTDIST on the output from SEQBOOT to compute the distance between each pair of sequences in each of the 10 replicated datasets.

c. Run NEIGHBOR (neighbor joining) on the output from PROTDIST to produce a tree for each replicate.

d. Run CONSENSE on the output from NEIGHBOR to find a single tree that best represents the 10 replicates, computing bootstrap values for each branch.

4) TreeView Version 1.6.6 was used to display and root the consensus tree.

In order to create a phylogenetic tree, you first must align the amino acid sequences.

Or you can bypass seeing the alignment and just view the trees directly.

If you are interested in creating other phylogenetic trees, you might try this ClustalW site at EBI.