Ska is a program for aligning and superposing proteins based on their structures. In contrast to sequence alignment, structural alignment depends solely on the geometric shape of the two protein chains being compared. That is, ska attempts to pair residues that are in geometrically equivalent positions in the 3D structures of two proteins, as opposed to pairing residues based on their identity and position in their linear primary sequences.
Click here to generate a license for ska. An e-mail will be sent to you once the license is accepted with a link to download the program. When you have downloaded the file follow the following instructions to install it:
- Create a directory for ska (e.g., "/location/of/ska") and move ska.tar.gz to that directory.
- cd to that directory and unpack ska with the command "tar -zxvf ska.tar.gz".
- Set the environment variables TROLLTOP=/location/of/ska/allh.top and SUBMAT=/location/of/ska/blosum62.
Instructions for running ska are below.
To run ska, you need two files, each containing a single protein structure in PDB (Protein Data Bank) format. Each file should contain only a single chain and other non-protein molecules such as ligands should be removed. For example, to align the structures of the A chains of human ERK2 and rat ERK2 from the PDB (PDB codes 1pme and 4n4s), you would create two new files (e.g. 1pme_A.pdb and 4n4s_A.pdb) containing only the A chains of the original PDB files (these files are provided in the ska download). The command syntax for ska is:
ska 1pme_A.pdb 4n4s_A.pdb
The main output of ska is a sequence alignment as shown below.
Lines beginning with tc_sse represent the secondary structure of the protein where "C" means loop, "E" means beta strand, and "H" means alpha helix. The other lines correspond to the primary sequence of the two protein chains. "RMSD" is the root mean square deviation in angstroms of the aligned residues' C-alpha atoms. "PSD" is the protein structural distance, another measure of protein structural similarity developed in the Honig lab. For both measures, the smaller the number, the more similar the two protein structures (for more information on PSD, see J. Mol. Biol. 301:665-78). You can obtain this alignment in fasta format by providing the "-f" option on the command line. You can obtain a PDB file containing the protein structures superposed according to this alignment using the "-o filename.pdb" option. The first four lines of this file contain the transformation that superposes the second protein on the first, based on the alignment. The rotation is in the first three lines and the translation is in the fourth.
Ska was developed in the Honig Lab and is supported by National Institutes of Health and National Science Foundation grants 4-R01-GM030518-37, S10-OD012351, S10 OD021764, and MCB-1412472.
If you have questions about using ska, contact email@example.com.