Multi-source Protein Signal Subcellular Localization Prediction And Optimization Design Of DNA Oligomers

Protein is an important component of all human cells and tissues.And plays an important role in human life activities.Proteins need to be transported to a specific subcellular location at the right time to perform their normal functions,otherwise it will have a significant impact on cellular function and even life.Protein subcellular localization prediction is one of the important ways to explore and reveal protein function,and it is also helpful for development of novel drug targets and the understanding of complex physiological processes.On the other hand,protein synthesis cannot be separated from the control of genetic material DNA,which is an essential biological macromolecule for development of organisms.Based on the law of DNA base complementarity,short DNA sequences,namely DNA oligomer sequences,can be used to construct nucleic acid transportation carriers that can safely enter and exit living organisms.Drugs,mi RNAs and other small molecular substances can be modified on the transport carriers at specific points to achieve the purpose of immune activation or therapy.Based on the above research motivation,the construction of protein subcellular localization prediction model and the construction of stable nucleic acid transportation carrier by optimizing the design of DNA oligomers were further studied in this paper.The main content of paper is summarized in two aspects as follows:(1)A method for subcellular localization prediction of multi-source protein signals based on error correction output coding strategy is proposed.Firstly,SVM(Support Vector Machine)is used to build three sequence-based classifiers and five image-based classifiers,then the results of the classifier are integrated with the error correction output coding,and the subcellular location of the protein is finally determined by encoding and decoding.Compared with the single protein image prediction method,the proposed method obtained satisfactory prediction results through the five-fold cross validation.The results show that the combination of image and sequence is more conducive to the prediction of protein subcellular localization.(2)An optimal design method of DNA oligomer based on simulated annealing algorithm was proposed.According to the proposed free energy of the objective function,the initialized sequence is iteratively optimized and updated.At the same time,the important constraints of DNA secondary structure and fragments are proposed.Taking tetramer as an example,the proposed algorithm ensures that the free energy of the target pairing region of DNA oligomer sequence is above-27kcal/mol and the melting temperature is above 52?.The free energy of the non-target pairing region between sequences can be reduced to within-7kcal/mol.The optimized results are verified by biological experiments,and the assembly effect between oligomer sequences is satisfactory,which proves the effectiveness of the algorithm.