RESUMEN

MOTIVATION: Protein-protein interaction sites (PPIS) are crucial for deciphering protein action mechanisms and related medical research, which is the key issue in protein action research. Recent studies have shown that graph neural networks have achieved outstanding performance in predicting PPIS. However, these studies often neglect the modeling of information at different scales in the graph and the symmetry of protein molecules within three-dimensional space. RESULTS: In response to this gap, this article proposes the MEG-PPIS approach, a PPIS prediction method based on multi-scale graph information and E(n) equivariant graph neural network (EGNN). There are two channels in MEG-PPIS: the original graph and the subgraph obtained by graph pooling. The model can iteratively update the features of the original graph and subgraph through the weight-sharing EGNN. Subsequently, the max-pooling operation aggregates the updated features of the original graph and subgraph. Ultimately, the model feeds node features into the prediction layer to obtain prediction results. Comparative assessments against other methods on benchmark datasets reveal that MEG-PPIS achieves optimal performance across all evaluation metrics and gets the fastest runtime. Furthermore, specific case studies demonstrate that our method can predict more true positive and true negative sites than the current best method, proving that our model achieves better performance in the PPIS prediction task. AVAILABILITY AND IMPLEMENTATION: The data and code are available at https://github.com/dhz234/MEG-PPIS.git.

Asunto(s)

Redes Neurales de la Computación , Mapeo de Interacción de Proteínas , Mapeo de Interacción de Proteínas/métodos , Proteínas/metabolismo , Proteínas/química , Algoritmos , Bases de Datos de Proteínas , Biología Computacional/métodos , Mapas de Interacción de Proteínas

RESUMEN

On the basis of the origin comparison of known endothelial genesis inhibitors, a 417-bp cDNA fragment was amplified from umbilical cord by RT-PCR and cloned into the expression vector pPIC9, followed by transformation into Pichia pastoris GS115. The resulted yeast was induced with methanol to express recombinant protein. The resulted protein was purified from culture broth and designated as EDI-8t. The in vitro study showed that EDI-8t, originated from collagen VIII, could specifically inhibit the growth and migration of bovine aortic endothelial cells (BAEC) stimulated by basic fibroblast growth factor (bFGF). The protein also exhibited the activity to cause cell apoptosis. In vivo EDI-8t showed the identical activity comparing with endostatin to inhibit the growth of liver tumor transplanted into nude mice. Interestingly, EDI-8t showed higher activity than endostatin to inhibit tumor growth in metastatic model of melanoma mice.

Asunto(s)

Inhibidores de la Angiogénesis/biosíntesis , Colágeno Tipo VIII/química , Pichia/metabolismo , Proteínas Recombinantes/farmacología , Secuencia de Aminoácidos , Inhibidores de la Angiogénesis/genética , Inhibidores de la Angiogénesis/aislamiento & purificación , Animales , Antineoplásicos/farmacología , Secuencia de Bases , Bovinos , Células Cultivadas , Colágeno Tipo VIII/genética , Endotelio Vascular/metabolismo , Vectores Genéticos/genética , Células Endoteliales de la Vena Umbilical Humana/química , Humanos , Ratones , Ratones Desnudos , Datos de Secuencia Molecular , Pichia/genética , Proteínas Recombinantes/biosíntesis , Proteínas Recombinantes/genética

RESUMEN

There is a highly homologous region in the C domain of the EGF family, some of its residues are semi-conserved. We constructed three hTGF-alpha mutants, hTGF-alphaV35, hTGF-alphaQ44, hTGF-alphaY45R46, by site-directed mutagenesis to replace the semi-conserved residues in the C domain of hTGF-alpha with the corresponding residues of hEGF. We observed that although the binding affinity of hEGF to hEGF receptor was about two fold that of hTGF-alpha, but the receptor binding affinity of the three mutants was respectively decreased to about 22 %, 13.4 % and 25 % compared of that of hTGF-alpha. On the other hand, the stimulating action of hEGF on NRK-49F cell proliferation was only 10 % that of hTGF-alpha, but those of the threemutants was about 4 fold, 10 fold and 5 fold more active than hTGF-alpha. Thus, the three mutants did not become more similar to hEGF in function. The functional difference between hEGF and hTGF-alpha was not simply determined by any single semi-conserved residue, but substitution at those sites in the C domain have altered the characters of hTGF-alpha sharply.