Superimposing structures
Last update on the 17th of December, 2019Here we find structures similar to MerA (1ZK7) and refine sequence alignment concerning structural alignment.
Structure similarity search
We used PDBeFold service with default parameters to find structures similar to 1ZK7. A lot of found hits were of good quality and extensive alignment length. We selected 3 structures corresponding to proteins with non mercuric oxidative functions. Their IDs and alignment details are shown in table 1.
PDB ID | Chain | Function | RMSD |
---|---|---|---|
4JQ9 | A | Dihydrolipoyl dehydrogenase | 1.1212 |
5V36 | B | Gluthation reductase | 0.9212 |
1ZDL | A | Thioredoxin reductase | 1.0815 |
These structures along with 1ZK7 (chain A) were submitted to multiple structural alignment and structure-based sequence alignment. One can download the resulted PDB file with superimposed structures (chain A: 1ZK7, chain B: 4JQ9, chain C: 5V36, chain D: 1ZDL) and structure-based alignment file. Superimposed structures are show in fig. 1.
![Superimposed structures.](./superimpose_full_view.png)
Structures are in fine agreement.
Structure-aware and structure-nonaware sequence alignments
Sequence alignment was obtained with MUSCLE with defaults. One can download the alignment file. SuiteMSA package was used to visualize two multiple alignments parallelwise. One observed discrepancy is shown in fig. 2. Structure-based alignment provides a shift of 1ZDL loop concerning the other sequences, whereas MUSCLE avoided introduction of gaps. Two alignments show 66.667% overall consistency.
![Discrepancy in alignments.](./loop_comparator.png)
Affected regions are show in table 2. Note that sequence alignment coordinates slightly differ from PDB sequence coordinates.
Structure | PDB coordinates | Sequence coordinates |
---|---|---|
1ZK7:A | 174-177 | 174-177 |
4JQ9:F | 173-176 | 174-177 |
5V36:B | 165-168 | 166-169 |
1ZDL:A | 216-219 | 179-182 |
Corresponding regions are protein loops so there might some uncertainty even at the structural level. However, quick glance at the structures favours structure-based alignment: 1ZDL loop is indeed shifted concerning the other structures (fig. 3).
![Alignment discrepancy.](./loop_superimpose_label.png)