Team:UESTC-Software/Proof Of Concept
<!DOCTYPE html>
Experimental Verification
Theoretical support
Our project is supported by a number of papers. Among them, Zhou Xiaoli [1] recombined two heat-resistant proteins, 1VE6 and 1jJI, to obtain proteins with better heat-resistant properties.
Both thermophilic esterase AFEST and apAPH are members of the α/β hydrolase folding superfamily. Their primary structure consistency is only 15%.
Their catalytic domains are similar(RMSD is only 2.6), but their substrate binding domain is completely different. The substrate binding domain of AFEST is cap structure, and apAPH's substrate binding domain is thruster structure. To study the structure-function relationship between these two enzymes, We decided to exchange the CAP domain of the AFEST and the propeller domain of the apAPH to construct the chimeric enzyme.
Then go through site-directed mutation PA->PAR / AA->AAM7
PAR proteins that have significantly improved heat resistance compared to other proteins , is formed by replacing domain 1ve6A02(in protein 1VE6) with domain 1jjiA00(in protein 1JJI). According to the 3DZD (3D Zernike Descriptors) algorithm:
The average inter-type Euclidean distance in the 100 classification we defined is 5. 8715.
Euclidean distance based on 3DZD between 1jjiA00 and 1ve6A02: 3. 785.
It can be seen that even though the sequence consistency is very low, the shape similarity is very high.
In the original text, the author judged whether the domain to be exchanged was similar according by the RMSD of spatial structure. However, in the face of a large amount of data, RMSD needed to translate and rotate aligned proteins for many times, so the calculation was too large that it was almost impossible to conduct systematic classification according to RMSD.
It can be seen that the primary structure and structure measured by RMSD limit the synthetic biologists' search and exploration for doamin with similar function. Based on the 3DZD and the database we built, synthetic biologists were able to write down domains that might be replaced based on the shape structure.
Comparison of classification effects
In order to verify the classification results we made after extracting the shape features of domain, we will classify the results, Comparison with existing database classifications
On the left is the existing CATH or SCOP type tag, and on the right is the domain of the same type in the CATH or SCOP type, and the proportion of the most enriched type in the 100 types after our clustering. (Number of domains in the type with the most existence in 0-100 types/Total number of CATH or Scop types)
The closer the similarity distribution is to 1, the more similar the two classifications are.
It can be seen that, compared with random distribution, even if we use stereo structure as the classification standard, which is different from the method of CATH/SCOP that evaluates sequence similarity first and then corrects manually, the similarity is still much higher than that of random distribution.
On the one hand, it reflects the correlation between protein sequence and spatial structure, on the other hand, it also proves that the results of this project classification can provide a consistent and even more reasonable explanation with the traditional explanation.
References
[1] Zhou Xiaoli . Recombination of protein domains to construct new functional enzymes [D]. Jilin University , 2012.