Homology based structure predictions

Homologous

Because we found no homologous structures in Task 2, we extended our list by using HHSearch.

We will use 13 proteins for creating a multiple alignment for homologous modeling. We choosen sequences to cover the whole protein and we payed specific attention on the transmembrane region.

PDB-ID	Identity	Description
1s79	37%	Kram
2yc6	32%	Kram
3p73	28%	Kram
1kcg	22%	Kram
1jfm	14%	Kram
1bii	22%	Kram
2p24	21%	Kram
1cd1	21%	Kram
2wy3	29%	Kram
1lqv	14%	Kram
3jts	25%	Kram
1ow0	22%	Kram
1hxm	18%	Kram
3lmy	19%	Kram

SwissModel

SwissProt is a server based tool provided by the SIB. It combines tools like PSI-PRED and DISOPRED for secondary structure and disordered region prediction.

The model created by SwissModel is based on a self hit, but we had no chance to exclude the protein itself from the prediction. Therefore we also run SwissModel in Alignment-Mode.(TODO)

Automated Mode

predicted model

Model information: Modelled residue range: 26 to 297
Based on template: 1a6zC (2.60 Å)
Sequence Identity [%]: 100
Evalue: 7.66e-163

Quality information: QMEAN Z-Score: -1.035

Estimated absolute model quality

Estimated density of model quality

Z-Score by category

predicted error

Even though the model is based on a self hit, the Z-Score is about -1, which means that the model is one standard deviation from the mean. The model is not quite unlikely but also not the most probable one.