Sequence-based mutation analysis BCKDHA protocol

Pymol

Notice that our BCKDHA protein contains a signal peptide at the beginning of the sequence, which was included in all the SNP predictions and the mutation analysis. For visualizing a PDB file was used, which does not contain information about the signal petide. The length of the signal peptide (45 aa) has to be considered, when selecting the mutation positions in pymol. (For all mutation positions 45 aa had to be subtracted). As one mutation (G29E) was chosen which occurs in the signal peptide, no pymol visualization of this mutation was possible.

The visualization of the mutated residues was accomplished following the tutorial in the pymol wiki([1]). Example for mutation Q80E:

• load PDB file '1U5B:A'
• zoom resi 80
• Wizard -> Mutagenesis
• select resi 80
• No mutation -> select resultant residue
• Click Apply, Done

back to Sequence-based mutation analysis BCKDHA

PSIBLAST

To obtain a pssm via psiblast we used the blastpgp tool. The following command returned the pssm for 5 iterations with psiblast:

blastpgp -i BCKDHA.fasta -d /data/blast/nr/nr -j 5 -Q pssm.mat -o psiblast_output.txt

Multiple Sequence Alignment

To find the homologue sequences to BCKDHA we used BLAST. We used the online tool[2]. After finishing the job we extracted the mammalien sequences by adding an "AND mammals" command in the search field. We used ClustalW to make a multiple Alignment of the homologue sequences and visualised it with Jalview.

SNAP

we used the command: snapfun -i BCKDHA.fasta -m mutations.txt -o SNAP.out
The 10 mutations have to be saved in the mutation.txt file.

SIFT

SIFT was run with default parameters (Selected databse to search: UniRef90 2011 Apr, Median conservation of sequences: 3.00, Remove sequences more then 90 percent identical to query.)