37 results about "Functional annotation" patented technology

The invention provides a method for analyzing data of prokaryotic proteogenomics rapidly and automatically. Users only need to provide mass spectral data and a corresponding database file and set a simple retrieval parameter; retrieval of data of proteogenomics can be completed; simultaneously, a user defined data retrieval result can also be compatible; therefore, the identification coverage rate of the data of proteogenomics is increased; in the method disclosed by the invention, library search engines of different algorithms are integrated in advance; disadvantages of a single retrieval method are made up; a user defined library search result can also be compatible; the method has good compatibility; the peptide fragment identification coverage rate is increased to the most extent; in the method disclosed by the invention, functional annotation of new genes is automatically completed; large-scale identification of protein post-translational modification and analysis of non-labelled quantitative proteomics are realized for the first time; and automatic and rapid deep analysis of the data of proteogenomics is really realized.

The invention belongs to the field of plant molecular biotechnology and gene engineering and particularly relates to a method for predicting functions of unknown genes of maize based on dynamic correlation between gene expression level and characters. The method comprises the following steps: firstly collecting grain transcripts obtained after pollinating maize inbred lines for 15 days, and sequencing to obtain gene expression level data; building a dynamic correlation analysis LA model; evaluating significance of LA; excavating dynamic correlation of gene expression patterns of whole genomes of the maize; and carrying out function annotation of genes of significant LA results and predicting functions of the genes. The method is capable of predicting the functions of the unknown genes by taking the phenomenon of dynamic correlation between co-expression patterns of gene pairs in the maize grains as sally port; compared with the conventional co-expression network construction, the dynamic correlation analysis is capable of rapidly finding regulation genes for regulating the co-expression patterns.

The invention belongs to the fields of plant molecular biotechnology and gene engineering and particularly relates to a method for exploring a lipid metabolism mechanism of corn kernels based on dynamic association analysis. The method specifically comprises the following steps: collecting gene expression quantity data of a corn selfing line; collecting gene data associated with oil content of the corn kernels; collecting oil content data of the kernels of association group formed by the corn selfing line; establishing a dynamic association analysis LA model; exploring a control gene for controlling dynamic change of a coexpression mode of 26 oil content associated genes in a whole genome range, and authenticating a candidate gene dynamically associated with the coexpression mode of the 26 oil content associated genes; carrying out function annotation on the candidate gene; detecting whether the candidate gene is in a QTl section of oil content characters; and carrying out section association analysis on oil content phenotype data of the corn selfing line. By taking the control of dynamic association of the coexpression mode of the known oil content associated genes as a key point, the control gene for controlling lipid metabolism of the kernels can be rapidly and effectively authenticated.

The invention discloses a deep learning method for DNA binding protein identification and functional annotation based on a self-attention mechanism. A data set is selected from a protein database, and the constructed deep learning model is trained and tested by using the selected data set, and then the trained deep learning model is used for training and testing. The deep learning model can predict whether protein can bind to DNA; wherein, the deep learning model includes coding layer, embedding layer, long short-term memory neural network layer (LSTM), convolutional neural network layer (CNN) and self-attention Layer (Self‑Attention). Through the deep learning model based on the self-attention mechanism, the present invention can predict whether it has the function of binding with DNA according to the primary sequence information of the protein, and find out the region with the binding function.

The invention discloses a lncRNA-protein association prediction method based on projected neighborhood non-negative matrix decomposition. First, according to the lncRNA sequence, lncRNA expression profile data, protein sequence, GO function annotation data of protein and the interaction network of lncRNA and protein, calculate A variety of lncRNA features, protein features, lncRNA similarity matrix and protein similarity matrix are obtained. Secondly, the integrated lncRNA similarity network is obtained by fusing multiple lncRNA similarity networks, and the integrated protein similarity network is obtained by fusing multiple protein similarity networks. Finally, combining the integrated lncRNA(protein) similarity network and multiple lncRNA(protein) features, a feature-projected neighborhood nonnegative matrix factorization algorithm is proposed to predict potential lncRNA-protein interactions. The present invention can not only accurately predict new lncRNA-protein interactions, but also predict new proteins (lncRNA) that are not associated with any lncRNA (protein), effectively avoiding high human and material resource consumption caused by biochemical experiments.