CN111667880A - Protein residue contact map prediction method based on depth residual error neural network - Google Patents
Protein residue contact map prediction method based on depth residual error neural network Download PDFInfo
- Publication number
- CN111667880A CN111667880A CN202010461169.2A CN202010461169A CN111667880A CN 111667880 A CN111667880 A CN 111667880A CN 202010461169 A CN202010461169 A CN 202010461169A CN 111667880 A CN111667880 A CN 111667880A
- Authority
- CN
- China
- Prior art keywords
- protein
- protein sequence
- neural network
- residual error
- contact map
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B15/00—ICT specially adapted for analysing two-dimensional or three-dimensional molecular structures, e.g. structural or functional relations or structure alignment
- G16B15/30—Drug targeting using structural data; Docking or binding prediction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B50/00—ICT programming tools or database systems specially adapted for bioinformatics
- G16B50/30—Data warehousing; Computing architectures
Abstract
A protein residue contact map prediction method based on a deep residual error neural network comprises the steps of firstly, inputting a protein sequence to be subjected to protein residue contact map prediction; then, converting the protein sequence into a 20 xL matrix by using a one-hot expression form of 20 common amino acids for the protein sequence, digitizing the protein sequence information, and calculating by using the 20 xL matrix to obtain a 20 xL covariance tensor, namely the characteristic of the input network; secondly, building a depth residual error neural network framework, collecting a protein sequence and a label of an existing protein contact label from a PDB library, calculating a characteristic tensor of the protein sequence, forming a data set with the corresponding label, and learning a prediction model on the data set by using the depth residual error neural network framework; and finally, inputting the protein characteristic tensor to be subjected to protein residue contact map prediction into a model to obtain the protein sequence residue contact map. The method has the advantages of low calculation cost and high prediction precision.
Description
Technical Field
The invention belongs to the fields of bioinformatics and computer application, and particularly relates to a protein residue contact map prediction method based on a deep residual error neural network.
Background
Proteins are one of the most biologically important macromolecules and have a wide range of functions, and the interaction between protein molecules is realized by contact interaction between partial residues, which is ubiquitous and indispensable in life activities. Therefore, the precise identification of the contact between protein residues is of great guiding significance in understanding the functions of proteins, analyzing the interrelation between biomolecules, designing new drugs, and the like.
The research literature found that many methods for predicting protein residue contact maps have been proposed, such as DNCON2(Adhikari, B., Hou, J.and Cheng, J. (2017) DNCON2: improved protein contact map prediction using two-level deep convolutional neural networks), NNcon (toggle, A.N., Wang, Z., Eickhold, J.and Cheng, J. (NN) 2009: improved protein contact map prediction using two-level deep convolutional neural networks), NNcon (toggle, A.N., Wang, Z., Eichhold, J.and Cheng, J. (NN) 2009: improved protein contact map prediction using two-level deep convolutional neural networks), two-dimensional contact map prediction using A-map, N.S. 2 d-dimensional neural networks, J.S. and Xin, J.S. Pair, N.S. 2. D-dimensional neural networks, J.S. Pair, and J.S. Pat. A.S. 7. prediction using two-dimensional neural networks, N.S. J.S. contact map prediction using A.S. J.S. A, N.S. 7, geiss B J, Ben-Hur a. predict interacting residues from sequence and structure). Although the existing method can be used for predicting the protein residue contact map, a large amount of training data sets and machine learning algorithms are generally used, so that the calculation cost is high, and meanwhile, the prediction accuracy cannot be guaranteed to be optimal due to the fact that noise information in the training sets is not paid enough attention.
In conclusion, the existing protein residue contact map prediction method has a great gap from the practical application requirements in the aspects of calculation cost and prediction precision, and needs to be improved urgently.
Disclosure of Invention
In order to overcome the defects of the existing protein residue contact map prediction method in two aspects of calculation cost and prediction precision, the invention provides a protein residue contact map prediction method based on a deep residual error neural network, which is low in calculation cost and high in prediction precision.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a method of protein residue contact map prediction based on a deep residual neural network, the method comprising the steps of:
1) inputting a protein sequence P with the length L to be subjected to residue contact map prediction;
2) the 20 common amino acids were represented using the one-hot coding scheme as follows:
‘A’：[1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘C’：[0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘D’：[0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘E’：[0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘F’：[0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘G’：[0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘H’：[0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘I’：[0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0]
‘K’：[0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0]
‘L’：[0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0]
‘M’：[0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0]
‘N’：[0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0]
‘P’：[0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0]
‘Q’：[0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0]
‘R’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0]
‘S’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0]
‘T’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0]
‘V’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0]
‘W’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0]
‘Y’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1]
3) the protein sequence P was converted into a 20 XL matrix, denoted M, using the one-hot representation of the 20 common amino acids;
4) calculate the average value of each row in M, and record as
Wherein, i is 1,2, …,20, MijIs the ith row and the jth column element in M;
5) calculate the standard deviation of each row in M, denoted as σi：
Wherein i is 1,2, …, 20;
6) each of the M data was normalized and designated βij：βij＝(Mij-mi)/σiWherein i is 1,2, …,20, j is 1,2, …, L;
7) using Sigmoid function, βijMapped between (0,1) and denoted as μij：
Wherein, i is 1,2, …,20, j is 1,2, …, L;
8) according to steps 4) to 7), M is converted into a matrix U ═ μ ═ Mij},i＝1,2,…,20，j＝1,2,…,L；
9) The matrix U is projected to a tensor size 20 × L × L, denoted Mfea：
Wherein, a is 1,2, …,20, i is 1,2, …, L, j is 1,2, …, L;
10) constructing a depth residual error neural network framework, wherein the depth residual error neural network framework consists of six parts, and the first part consists of a convolution layer, a normalization layer and a ReLU layer; the second, third, fourth and fifth parts are composed of the same residual block, and the residual block is provided with two convolution layers, a normalization layer and a ReLU layer; the sixth part consists of a convolution layer and a Sigmoid function;
11) the protein sequence of the existing protein contact tag was collected from the PDB library and designated as Dataset ═ Pi,YiIn which P isiDenotes the i-th bar in the protein sequence, YiRepresents Pi1,2, …, N being the total number of protein sequences;
12) according to steps 1) to 9), generating all PiIs/are as follows
And a corresponding label YiComposing a sample set
13) Learning a prediction model on S by using the deep residual error neural network frame built in the step 10), and recording as DRN;
14) m of the protein P to be detectedfeaInputting the model DRN to obtain a residue contact map of the protein sequence.
The technical conception of the invention is as follows: firstly, inputting a protein sequence to be subjected to protein residue contact map prediction; then, for the protein sequence, the protein sequence was converted into a 20 × L matrix using one-hot representation of 20 common amino acids for the purpose of digitizing the protein sequence information; by utilizing a 20 multiplied by L matrix, a covariance tensor of 20 multiplied by L is obtained through calculation, namely the characteristic of the input network; secondly, building a depth residual error neural network framework, collecting a protein sequence and a label of an existing protein contact label from a PDB library, calculating a characteristic tensor of the protein sequence, forming a data set with the corresponding label, and learning a prediction model on the data set by using the depth residual error neural network framework; and finally, inputting the protein characteristic tensor to be subjected to protein residue contact map prediction into a model to obtain the protein sequence residue contact map. The invention provides a protein residue contact map prediction method based on a deep residual error neural network, which is low in calculation cost and high in prediction precision.
The beneficial effects of the invention are as follows: on one hand, the characteristic information of residues is extracted from the protein sequence information, the contact probability between the residues of the protein to be predicted is directly calculated, and the prediction efficiency of binding the residues by the protein ligand is improved; on the other hand, a multidimensional covariance feature matrix is introduced to express the correlation of contact action between residues, and the prediction precision of protein residue contact is improved.
Drawings
FIG. 1 is a diagram of a protein residue contact map prediction method based on a deep residual neural network.
FIG. 2 is the results of prediction of the contact pattern of residues T0759-D1 of the protein using a protein residue contact pattern prediction method based on a deep residual neural network.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
Referring to fig. 1 and 2, a protein residue contact map prediction method based on a deep residual neural network includes the following steps:
1) inputting a protein sequence P with the length L to be subjected to residue contact map prediction;
2) the 20 common amino acids were represented using the one-hot coding scheme as follows:
‘A’：[1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘C’：[0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘D’：[0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘E’：[0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘F’：[0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘G’：[0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘H’：[0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘I’：[0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0]
‘K’：[0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0]
‘L’：[0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0]
‘M’：[0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0]
‘N’：[0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0]
‘P’：[0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0]
‘Q’：[0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0]
‘R’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0]
‘S’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0]
‘T’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0]
‘V’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0]
‘W’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0]
‘Y’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1]
3) the protein sequence P was converted into a 20 XL matrix, denoted M, using the one-hot representation of the 20 common amino acids;
4) calculate the average value of each row in M, and record as
Wherein, i is 1,2, …,20, MijIs the ith row and the jth column element in M;
5) calculate the standard deviation of each row in M, denoted as σi：
Wherein i is 1,2, …, 20;
6) each of the M data was normalized and designated βij：βij＝(Mij-mi)/σiWherein i is 1,2, …,20, j is 1,2, …, L;
7) using Sigmoid function, βijMapped between (0,1) and denoted as μij：
Wherein, i is 1,2, …,20, j is 1,2, …, L;
8) according to steps 4) to 7), M is converted into a matrix U ═ μ ═ Mij},i＝1,2,…,20，j＝1,2,…,L；
9) The matrix U is projected to a tensor size 20 × L × L, denoted Mfea：
Wherein, a is 1,2, …,20, i is 1,2, …, L, j is 1,2, …, L;
10) constructing a depth residual error neural network framework, wherein the depth residual error neural network framework consists of six parts, and the first part consists of a convolution layer, a normalization layer and a ReLU layer; the second, third, fourth and fifth parts are composed of the same residual block, and the residual block is provided with two convolution layers, a normalization layer and a ReLU layer; the sixth part consists of a convolution layer and a Sigmoid function;
11) the protein sequence of the existing protein contact tag was collected from the PDB library and designated as Dataset ═ Pi,YiIn which P isiDenotes the i-th bar in the protein sequence, YiRepresents Pi1,2, …, N being the total number of protein sequences;
12) according to steps 1) to 9), generating all PiIs/are as follows
And a corresponding label YiComposing a sample set
13) Learning a prediction model on S by using the deep residual error neural network frame built in the step 10), and recording as DRN;
14) the protein to be testedM of PfeaInputting the model DRN to obtain a residue contact map of the protein sequence.
Using the residue contact map prediction of the protein T0759-D1 sequence as an example, the residue contact map of the protein T0759-D1 sequence predicted using the above method is shown in FIG. 2.
The above description is the prediction result obtained by the prediction of the residue contact map of the protein T0759-D1 sequence as an example, and is not intended to limit the scope of the present invention, and various modifications and improvements can be made without departing from the scope of the present invention.
Claims (1)
1. A protein residue contact map prediction method based on a deep residual error neural network is characterized by comprising the following steps:
1) inputting a protein sequence P with the length L to be subjected to residue contact map prediction;
2) the 20 common amino acids were represented using the one-hot coding scheme as follows:
‘A’：[1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘C’：[0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘D’：[0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘E’：[0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘F’：[0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘G’：[0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘H’：[0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0，0]
‘I’：[0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0，0]
‘K’：[0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0，0]
‘L’：[0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0，0]
‘M’：[0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0，0]
‘N’：[0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0，0]
‘P’：[0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0，0]
‘Q’：[0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0，0]
‘R’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0，0]
‘S’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0，0]
‘T’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0，0]
‘V’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0，0]
‘W’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1，0]
‘Y’：[0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，0，1]
3) the protein sequence P was converted into a 20 XL matrix, denoted M, using the one-hot representation of the 20 common amino acids;
4) calculate the average value of each row in M, and record as Mi：
Wherein, i is 1,2, …,20, MijIs the ith row and the jth column element in M;
5) calculate the standard deviation of each row in M, denoted as σi：
Wherein i is 1,2, …, 20;
6) each of the M data was normalized and designated βij：βij＝(Mij-mi)/σiWherein i is 1,2, …,20, j is 1,2, …, L;
7) using Sigmoid function, βijMapped between (0,1) and denoted as μij：
Wherein, i is 1,2, …,20, j is 1,2, …, L;
8) according to steps 4) to 7), M is converted into a matrix U ═ μ ═ Mij},i＝1,2,…,20，j＝1,2,…,L；
9) The matrix U is projected to a tensor size 20 × L × L, denoted Mfea：
Wherein, a is 1,2, …,20, i is 1,2, …, L, j is 1,2, …, L;
10) constructing a depth residual error neural network framework, wherein the depth residual error neural network framework consists of six parts, and the first part consists of a convolution layer, a normalization layer and a ReLU layer; the second, third, fourth and fifth parts are composed of the same residual block, and the residual block is provided with two convolution layers, a normalization layer and a ReLU layer; the sixth part consists of a convolution layer and a Sigmoid function;
11) the protein sequence of the existing protein contact tag was collected from the PDB library and designated as Dataset ═ Pi,YiIn which P isiDenotes the i-th bar in the protein sequence, YiRepresents Pi1,2, …, N being the total number of protein sequences;
12) according to steps 1) to 9), generating all PiIs/are as follows
And a corresponding label YiComposing a sample set
13) Learning a prediction model on S by using the deep residual error neural network frame built in the step 10), and recording as DRN;
14) m of the protein P to be detectedfeaInputting the model DRN to obtain a residue contact map of the protein sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010461169.2A CN111667880A (en) | 2020-05-27 | 2020-05-27 | Protein residue contact map prediction method based on depth residual error neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010461169.2A CN111667880A (en) | 2020-05-27 | 2020-05-27 | Protein residue contact map prediction method based on depth residual error neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111667880A true CN111667880A (en) | 2020-09-15 |
Family
ID=72384704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010461169.2A Pending CN111667880A (en) | 2020-05-27 | 2020-05-27 | Protein residue contact map prediction method based on depth residual error neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111667880A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112185466A (en) * | 2020-09-24 | 2021-01-05 | 中国科学院计算技术研究所 | Method for constructing protein structure by directly utilizing protein multi-sequence association information |
| CN112837740A (en) * | 2021-01-21 | 2021-05-25 | 浙江工业大学 | DNA binding residue prediction method based on structural features |
| CN112837741A (en) * | 2021-01-25 | 2021-05-25 | 浙江工业大学 | Protein secondary structure prediction method based on cyclic neural network |
| CN113096722A (en) * | 2021-03-17 | 2021-07-09 | 浙江工业大学 | Protein phosphorylation site prediction method based on inner product self-attention neural network |
| CN113571126A (en) * | 2021-07-06 | 2021-10-29 | 浙江工业大学 | Protein residue contact prediction method based on multi-loss training |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231677A (en) * | 2007-11-30 | 2008-07-30 | 中国科学院合肥物质科学研究院 | Long-distance interaction prediction method between residue base on sequence spectrum center and genetic optimization process |
| CN107622182A (en) * | 2017-08-04 | 2018-01-23 | 中南大学 | The Forecasting Methodology and system of protein partial structurtes feature |
| CN108763865A (en) * | 2018-05-21 | 2018-11-06 | 成都信息工程大学 | A kind of integrated learning approach of prediction DNA protein binding sites |
| CN110070909A (en) * | 2019-03-21 | 2019-07-30 | 中南大学 | A kind of protein function prediction technique of the fusion multiple features based on deep learning |
| CN110390995A (en) * | 2019-07-01 | 2019-10-29 | 上海交通大学 | α spiral transmembrane protein topological structure prediction technique and device |
| CN110870019A (en) * | 2017-10-16 | 2020-03-06 | 因美纳有限公司 | Semi-supervised learning for training deep convolutional neural network sets |
| CN111063389A (en) * | 2019-12-04 | 2020-04-24 | 浙江工业大学 | Ligand binding residue prediction method based on deep convolutional neural network |
| CN111081312A (en) * | 2019-12-04 | 2020-04-28 | 浙江工业大学 | Ligand binding residue prediction method based on multi-sequence association information |
-
2020
- 2020-05-27 CN CN202010461169.2A patent/CN111667880A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231677A (en) * | 2007-11-30 | 2008-07-30 | 中国科学院合肥物质科学研究院 | Long-distance interaction prediction method between residue base on sequence spectrum center and genetic optimization process |
| CN107622182A (en) * | 2017-08-04 | 2018-01-23 | 中南大学 | The Forecasting Methodology and system of protein partial structurtes feature |
| CN110870019A (en) * | 2017-10-16 | 2020-03-06 | 因美纳有限公司 | Semi-supervised learning for training deep convolutional neural network sets |
| CN108763865A (en) * | 2018-05-21 | 2018-11-06 | 成都信息工程大学 | A kind of integrated learning approach of prediction DNA protein binding sites |
| CN110070909A (en) * | 2019-03-21 | 2019-07-30 | 中南大学 | A kind of protein function prediction technique of the fusion multiple features based on deep learning |
| CN110390995A (en) * | 2019-07-01 | 2019-10-29 | 上海交通大学 | α spiral transmembrane protein topological structure prediction technique and device |
| CN111063389A (en) * | 2019-12-04 | 2020-04-24 | 浙江工业大学 | Ligand binding residue prediction method based on deep convolutional neural network |
| CN111081312A (en) * | 2019-12-04 | 2020-04-28 | 浙江工业大学 | Ligand binding residue prediction method based on multi-sequence association information |
Non-Patent Citations (5)
| Title |
|---|
| QIWANG 等: ""A novel framework for the identification of drug target proteins:Combining stacked auto-encoders with a biased support vector machine"", 《PLOS ONE》 * |
| WXYSTYLE: ""机器学习：数据预处理之独热编码（One-Hot）"", 《HTTPS://BLOG.CSDN.NET/WXYSTYLE/ARTICLE/DETAILS/80729741》 * |
| XIAOYANG JING 等: ""Amino Acid Encoding Methods for Protein Sequences: A Comprehensive Review and Assessment"", 《IEEE/ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS》 * |
| ZHONG LI 等: ""Protein Contact Map Prediction Based on ResNet and DenseNet"", 《HINDAWI BIOMED RESEARCH INTERNATIONA》 * |
| 於东军 等: ""蛋白质残基接触图预测"", 《南京理工大学学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112185466A (en) * | 2020-09-24 | 2021-01-05 | 中国科学院计算技术研究所 | Method for constructing protein structure by directly utilizing protein multi-sequence association information |
| CN112185466B (en) * | 2020-09-24 | 2023-05-23 | 中国科学院计算技术研究所 | Method for constructing protein structure by directly utilizing protein multi-sequence association information |
| CN112837740A (en) * | 2021-01-21 | 2021-05-25 | 浙江工业大学 | DNA binding residue prediction method based on structural features |
| CN112837740B (en) * | 2021-01-21 | 2024-03-26 | 浙江工业大学 | DNA binding residue prediction method based on structural characteristics |
| CN112837741A (en) * | 2021-01-25 | 2021-05-25 | 浙江工业大学 | Protein secondary structure prediction method based on cyclic neural network |
| CN112837741B (en) * | 2021-01-25 | 2024-04-16 | 浙江工业大学 | Protein secondary structure prediction method based on cyclic neural network |
| CN113096722A (en) * | 2021-03-17 | 2021-07-09 | 浙江工业大学 | Protein phosphorylation site prediction method based on inner product self-attention neural network |
| CN113571126A (en) * | 2021-07-06 | 2021-10-29 | 浙江工业大学 | Protein residue contact prediction method based on multi-loss training |
| CN113571126B (en) * | 2021-07-06 | 2024-03-22 | 浙江工业大学 | Protein residue contact prediction method based on multi-loss training |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111667880A (en) | Protein residue contact map prediction method based on depth residual error neural network | |
| Lan et al. | A survey of data mining and deep learning in bioinformatics | |
| Zheng et al. | SinNLRR: a robust subspace clustering method for cell type detection by non-negative and low-rank representation | |
| Lin et al. | scJoint integrates atlas-scale single-cell RNA-seq and ATAC-seq data with transfer learning | |
| Zhao et al. | HyperAttentionDTI: improving drug–protein interaction prediction by sequence-based deep learning with attention mechanism | |
| CN108334574B (en) | Cross-modal retrieval method based on collaborative matrix decomposition | |
| CN113707235B (en) | Drug micromolecule property prediction method, device and equipment based on self-supervision learning | |
| Chen et al. | A novel information-theoretic approach for variable clustering and predictive modeling using Dirichlet process mixtures | |
| Wang et al. | Molecular property prediction based on a multichannel substructure graph | |
| CN113393911A (en) | Ligand compound rapid pre-screening model based on deep learning | |
| Erfanian et al. | Deep learning applications in single-cell omics data analysis | |
| Nasab et al. | Deep learning in spatially resolved transcriptomics: a comprehensive technical view | |
| Durugkar et al. | Introduction to data mining | |
| Bellazzi et al. | The Gene Mover's Distance: Single-cell similarity via Optimal Transport | |
| Vengatesan et al. | The performance analysis of microarray data using occurrence clustering | |
| CN109857892B (en) | Semi-supervised cross-modal Hash retrieval method based on class label transfer | |
| Ngo et al. | Multiresolution graph transformers and wavelet positional encoding for learning long-range and hierarchical structures | |
| CN113724195A (en) | Protein quantitative analysis model based on immunofluorescence image and establishment method | |
| CN113160886B (en) | Cell type prediction system based on single cell Hi-C data | |
| CN113361752B (en) | Protein solvent accessibility prediction method based on multi-view learning | |
| CN113257357B (en) | Protein residue contact map prediction method | |
| Qu et al. | Enhancing understandability of omics data with shap, embedding projections and interactive visualisations | |
| Zhen et al. | A novel framework for single-cell hi-c clustering based on graph-convolution-based imputation and two-phase-based feature extraction | |
| JP2024513995A (en) | Multichannel protein voxelization to predict variant pathogenicity using deep convolutional neural networks | |
| Kumbhar et al. | Web mining: A Synergic approach resorting to classifications and clustering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200915 |