CN105808972A - Method for predicting protein structure from local to global on basis of knowledge spectrum - Google Patents
Method for predicting protein structure from local to global on basis of knowledge spectrum Download PDFInfo
- Publication number
- CN105808972A CN105808972A CN201610139514.4A CN201610139514A CN105808972A CN 105808972 A CN105808972 A CN 105808972A CN 201610139514 A CN201610139514 A CN 201610139514A CN 105808972 A CN105808972 A CN 105808972A
- Authority
- CN
- China
- Prior art keywords
- distance
- sequence
- protein
- spectrum
- search sequence
- 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
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
Landscapes
- Spectroscopy & Molecular Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biotechnology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Crystallography & Structural Chemistry (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a method for predicting a protein structure from local to global on the basis of a knowledge spectrum. The method comprises the following steps: for a query sequence, obtaining a high-quality fragment library through a multi-feature seamless threading method, and obtaining distance spectrum knowledge between residues through statistic consistency analysis on the basis of the fragment library; dividing the query sequence into several segments of structures according to residue information recorded in a distance spectrum; carrying out fragment assembling on each segment of structure to obtain relatively low energy and to ensure that the space distance between adjacent residues approaches to a predicted distance in the distance spectrum; carrying out fragment assembling on the unsegmented structures and calculating the global energy to obtain a metastable-state conformation with low energy and more reasonable structure. The method disclosed in the invention is relatively good in conformation space sampling ability and high in prediction precision.
Description
Technical field
The present invention relates to bioinformatics, computer application field, in particular a kind of based on spectrum knowledge from
Local is to the Advances in protein structure prediction of the overall situation.
Background technology
Protein molecule plays vital effect in biological cell chemical reaction process.Their structure mould
Type and biological activity state are to we have appreciated that and cure multiple disease have important meaning.Protein is only folded into
Specific three dimensional structure could produce its distinctive biological function.It is therefore to be understood that the function of protein, just
Its three-D space structure must be obtained.
Tertiary protein structure prediction is a vital task of bioinformatics.Protein conformation optimization problem is existing
It is that extremely complex protein energy function surface is scanned for facing maximum challenge.Protein energy mould
Type considers molecular system bonding effect and Van der Waals force, electrostatic, hydrogen bond, the non-bonding effect such as hydrophobic, causes
The Energy Surface formed it into is the most coarse, and conformation correspondence local minimizers number number is finger with the increase of sequence length
Number increases.And the mechanism that protein conformation prediction algorithm can find protein stabilization structure is, substantial amounts of albumen
Matter meta structure constitutes low energy area, thus can find protein the overall situation rock-steady structure it is crucial that
Algorithm can find substantial amounts of protein meta structure, i.e. increases the population diversity of algorithm.Therefore, for
More accurate protein force field model, chooses effective conformational space optimized algorithm, makes new protein structure
Prediction algorithm has more universality and high efficiency becomes the focal issue of protein structure prediction in bioinformatics.
At present, Advances in protein structure prediction substantially can be divided into two classes, method based on template and be not based on mould
The method of plate.Wherein, ab initio prediction (Ab-inito) method being not based on template is most widely used.It is suitable for
In the homology most protein less than 25%, only produce brand new from sequence, to Protein Molecular Design
And the research of protein folding etc. is significant.Currently there is following several comparison successful ab initio prediction method:
Zhang Yangyu Jeffrey Skolnick cooperation TASSER (Threading/Assembly/Refinement) method,
The FeLTr method etc. of the designs such as David Baker and the Rosetta method of team's design, Shehu.But up till now
Also not having a kind of the most perfect method to carry out the three dimensional structure of predicted protein matter, well predicting knot even if obtaining
Really, but also just for some protein for, technical bottleneck currently mainly is two aspects, first
Aspect is the method for sampling, and prior art is not strong to conformational space ability in sampling, further aspect is that conformation updates
Method, prior art is the most not enough to the renewal precision of conformation.
Therefore, existing Advances in protein structure prediction Shortcomings, need to improve.
Summary of the invention
In order to overcome that the conformational space ability in sampling of existing Advances in protein structure prediction is strong, precision of prediction relatively
Low deficiency, the present invention proposes that a kind of conformational space ability in sampling is preferable, precision of prediction is high based on spectrum knowledge from
Local is to the Advances in protein structure prediction of the overall situation.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of based on spectrum knowledge from local to the overall situation Advances in protein structure prediction, described optimization method include with
Lower step:
1) given search sequence information；
2) from Protein Data Bank (PDB) website to be, download resolution less thanHigh accuracy protein, wherein
For parasang,Rice, removes sequence similarity according to sequence alignment algorithms NW-Align and is more than
The amino acid chain of 30%, obtains non-redundant proteins template base；
3) according to multi-feature similarity function:
By the protein chain in seamless threading method comparison nonredundancy template base relative to search sequence each residue position
(i, j), wherein i is search sequence resi-dues to score f put, and j is fragment structure；F (i, j) in, under
Mark q represents search sequence feature score item, and subscript t represents template protein feature score item, Pq(i,k)
The sequence frequency obtained by PSI-BLAST for search sequence is composed, and wherein k is predetermined number amino acids
Type；Lq(i, k) and Lt(j k) is the search sequence obtained by PSI-BLAST and template sequence logarithmic spectrum；
sstJ () is the classification of template protein secondary structure, DSSP be calculated；ssqI () is search sequence two grades
Textural classification, can be obtained by two layers of neural metwork training；sat(j) and saqI () is formwork structure and search sequence
Solvent accessibility index, is trained by EDTSurf and neural network procedure and obtains；ψqJ () is inquiry
Sequence dihedral angle is to being obtained by two layers of neural metwork training；ψtJ () passes through query protein word
Allusion quotation obtains；SPt(j k) is the structure spectrum of template protein；w1、w2、w3、w4And w5For weighted value；
4) according to similarity score f, (i j) chooses M fragment of highest scoring on each position of search sequence and obtains fragment
Library file；
5) statistical query sequence is to the distance between coming from a template segments, and the most only statistics is less than's
Residue to spacing, draw rectangular histogram and obtain distance spectrum, the distance of rectangular histogram abscissa is spaced apart
When in template residue between distance in certain interval, then this interval sum just adds 1, if broken line graph existsThere is peak value in certain interior distance interval, then the distance interval that this peak value is corresponding is in target sequence
Residue i, to the Prediction distance of residue j, records the distance spectrum (profile) that this distribution is between two residues；
6) according to the position of residue in gained distance spectrum, search sequence is divided into n section；
7) l=1 is made, l ∈ 1,2,3 ..., and n}, operation following to segmental structure execution:
7.1) l section fragment structure is carried out fragment assembling；
7.2) calculate the distance between the resi-dues wherein containing distance spectrum information, and seek deviation with Prediction distance,
Cumulative deviation value is also averaged and is designated as Δ D；
7.3) if Δ D < R, then this conformation storage being designated as cell, wherein R is structural precision constraints；
7.4) 7.1 are repeated) to 7.3) until storage cell number reaches 100 cell, based on Rosetta Score 3
The energy of segmental structure in comparison cell, the structure choosing minimum energy is the prediction knot of this segmentation
Structure；
7.5) l=l+1, it is judged that whether l, more than or equal to n, is to enter 8), otherwise return to 7.1)；
8) arranging iterations is G, makes s=1, the following operation of execution:
8.1) target induced conformational ENERGY E (P is calculatedtarget)；
8.2) unsegmented structure is carried out fragment assembling, calculate energy value E (Ptrail)；
8.3) if E is (Ptarget)>E(Ptrail) then use PtrailReplace Ptarget；
8.4) s=s+1；Judge that s, whether more than or equal to G, is to enter 9), otherwise return to 8.1)；
9) output induced conformational, obtains search sequence nearly native state structure.
The technology of the present invention is contemplated that: obtain high-quality firstly, for search sequence by the seamless threading method of multiple features
Amount fragment library, obtains the distance spectrum knowledge between residue-residue based on fragment library by statistics consistency analysis；Then,
Search sequence is divided into and is divided into a few segment structure according to the residue information of record in distance spectrum；Afterwards, for each section
Structure is assembled by fragment and obtains that energy is relatively low and space length between residue-residue approaches in distance spectrum and finds range in advance
From；Carry out fragment assembling finally, for unsegmented structure, calculate global energy, obtain that energy is low and structure more
For rational metastable state conformation.
The invention have the benefit that conformational space ability in sampling is preferable, precision of prediction is high.Accompanying drawing explanation
Fig. 1 is cycle tests relation schematic diagram of RMSD and energy value during population recruitment.
Fig. 2 is 1ENH algorithm predicts gained protein three-dimensional structure schematic diagram.
Detailed description of the invention
The invention will be further described below in conjunction with the accompanying drawings.
With reference to Fig. 1～Fig. 2, a kind of based on spectrum knowledge from local to the Advances in protein structure prediction of the overall situation, including
Following steps:
1) given search sequence information；
2) from Protein Data Bank (PDB) website to be, download resolution less thanHigh accuracy protein, wherein
For parasang,Rice, removes sequence similarity according to sequence alignment algorithms NW-Align and is more than
The amino acid chain of 30%, obtains non-redundant proteins template base；
3) according to multi-feature similarity function:
By the protein chain in seamless threading method comparison nonredundancy template base relative to search sequence each residue position
(i, j), wherein i is search sequence resi-dues to score f put, and j is fragment structure；F (i, j) in, under
Mark q represents search sequence feature score item, and subscript t represents template protein feature score item, Pq(i,k)
The sequence frequency obtained by PSI-BLAST for search sequence is composed, and wherein k is predetermined number amino acids
Type；Lq(i, k) and Lt(j k) is the search sequence obtained by PSI-BLAST and template sequence logarithmic spectrum；
sstJ () is the classification of template protein secondary structure, DSSP be calculated；ssqI () is search sequence two grades
Textural classification, can be obtained by two layers of neural metwork training；sat(j) and saqI () is formwork structure and search sequence
Solvent accessibility index, is trained by EDTSurf and neural network procedure and obtains；ψqJ () is inquiry
Sequence dihedral angle is to being obtained by two layers of neural metwork training；ψtJ () passes through query protein word
Allusion quotation obtains；SPt(j k) is the structure spectrum of template protein；w1、w2、w3、w4And w5For weighted value；
4) according to similarity score f, (i j) chooses M fragment of highest scoring on each position of search sequence and obtains fragment
Library file；
5) statistical query sequence is to the distance between coming from a template segments, and the most only statistics is less than's
Residue to spacing, draw rectangular histogram and obtain distance spectrum, the distance of rectangular histogram abscissa is spaced apart
When in template residue between distance in certain interval, then this interval sum just adds 1, if broken line graph existsThere is peak value in certain interior distance interval, then the distance interval that this peak value is corresponding is in target sequence
Residue i, to the Prediction distance of residue j, records the distance spectrum (profile) that this distribution is between two residues；
6) according to the position of residue in gained distance spectrum, search sequence is divided into n section；
7) l=1 is made, l ∈ 1,2,3 ..., and n}, operation following to segmental structure execution:
7.1) l section fragment structure is carried out fragment assembling；
7.2) calculate the distance between the resi-dues wherein containing distance spectrum information, and seek deviation with Prediction distance,
Cumulative deviation value is also averaged and is designated as Δ D；
7.3) if Δ D < R, then this conformation storage being designated as cell, wherein R is structural precision constraints；
7.4) 7.1 are repeated) to 7.3), until storage cell number reaches x cell, compare based on Rosetta Score 3
To the energy of segmental structure in cell, the structure choosing minimum energy is the pre-geodesic structure of this segmentation；
7.5) l=l+1, it is judged that whether l, more than or equal to n, is to enter 8), otherwise return to 7.1)；
8) arranging iterations is G, makes s=1, the following operation of execution:
8.1) calculate target induced conformational ENERGY E (Ptarget)；
8.2) unsegmented structure is carried out fragment assembling, calculate energy value E (Ptrail)；
8.3) if E is (Ptarget)>E(Ptrail) then use PtrailReplace Ptarget；
8.4) s=s+1；Judge that s, whether more than or equal to G, is to enter 9), otherwise return to 8.1)；
9) output induced conformational, obtains search sequence nearly native state structure.
The present embodiment with protein 1ENH that sequence length is 54 as embodiment, a kind of based on spectrum knowledge from local
To the Advances in protein structure prediction of the overall situation, wherein comprise the steps of
1) given search sequence information；
2) from Protein Data Bank (PDB) website to be, download resolution less thanHigh accuracy protein, wherein
For parasang,Rice, removes sequence similarity according to sequence alignment algorithms NW-Align and is more than
The amino acid chain of 30%, obtains 8619 protein chains and constitutes nonredundancy template base；
3) according to multi-feature similarity function:
By the protein chain in seamless threading method comparison nonredundancy template base relative to search sequence each residue position
(i, j), wherein i is search sequence resi-dues to score f put, and j is fragment structure；F (i, j) in, under
Mark q represents search sequence feature score item, and subscript t represents template protein feature score item, Pq(i,k)
The sequence frequency obtained by PSI-BLAST for search sequence is composed, and wherein k is predetermined number amino acids
Type；Lq(i, k) and Lt(j k) is the search sequence obtained by PSI-BLAST and template sequence logarithmic spectrum；
sstJ () is the classification of template protein secondary structure, DSSP be calculated；ssqI () is search sequence two grades
Textural classification, can be obtained by two layers of neural metwork training；sat(j) and saqI () is formwork structure and search sequence
Solvent accessibility index, is trained by EDTSurf and neural network procedure and obtains；ψqJ () is inquiry
Sequence dihedral angle is to being obtained by two layers of neural metwork training；ψtJ () passes through query protein word
Allusion quotation obtains；SPt(j k) is the structure spectrum of template protein；w1、w2、w3、w4And w5For weighted value；
4) according to similarity score f, (i, 200 fragments j) choosing highest scoring on each position of search sequence obtain sheet
Phase library file；
5) statistical query sequence is to the distance between coming from a template segments, and the most only statistics is less than's
Residue to spacing, draw rectangular histogram and obtain distance spectrum, the distance of rectangular histogram abscissa is spaced apart
When in template residue between distance in certain interval, then this interval sum just adds 1, if broken line graph existsThere is peak value in certain interior distance interval, then the distance interval that this peak value is corresponding is in target sequence
Residue i, to the Prediction distance of residue j, records the distance spectrum (profile) that this distribution is between two residues；
6) according to the position of residue in gained distance spectrum, search sequence is divided into n section；
7) l=1 is made, l ∈ 1,2,3 ..., and n}, operation following to segmental structure execution:
7.1) l section fragment structure is carried out fragment assembling；
7.2) calculate the distance between the resi-dues wherein containing distance spectrum information, and seek deviation with Prediction distance,
Cumulative deviation value is also averaged and is designated as Δ D；
7.3) if Δ D < R, then this conformation storage being designated as cell, wherein R is structural precision constraints；
7.4) 7.1 are repeated) to 7.3), until storage cell number reaches x cell, compare based on Rosetta Score 3
To the energy of segmental structure in cell, the structure choosing minimum energy is the pre-geodesic structure of this segmentation；
7.5) l=l+1, it is judged that whether l, more than or equal to n, is to enter 8), otherwise return to 7.1)；
8) arranging iterations is G=50000, makes s=1, the following operation of execution:
8.1) target induced conformational ENERGY E (P is calculatedtarget)；
8.2) unsegmented structure is carried out fragment assembling, calculate energy value E (Ptrail)；
8.3) if E is (Ptarget)>E(Ptrail) then use PtrailReplace Ptarget；
8.4) s=s+1；Judge that s, whether more than or equal to G=50000, is to enter 9), otherwise return to 8.1)；
9) output induced conformational, obtains search sequence nearly native state structure.
With protein 1ENH that sequence length is 54 as embodiment, above method is used to obtain this protein
Nearly native state conformation, in conformation assemblage, conformation updates figure as it is shown in figure 1, algorithm predicts gained protein tridimensional is tied
Structure is shown as shown in Figure 2.
The excellent results that the embodiment that the present invention is given that described above is shows, it is clear that the present invention is not
Be suitable only for above-described embodiment, without departing from essence spirit of the present invention and without departing from involved by flesh and blood of the present invention in
It can be done many variations on the premise of appearance to be carried out.
Claims (1)
1. one kind based on spectrum knowledge from local to the overall situation Advances in protein structure prediction, it is characterised in that: described albumen
Matter Structure Prediction Methods comprises the following steps:
1) given search sequence information；
2) from Protein Data Bank website to be, download resolution less thanHigh accuracy protein, whereinFor distance
Unit,Remove sequence similarity according to sequence alignment algorithms NW-Align and be more than 30%
Amino acid chain, obtains non-redundant proteins template base；
3) according to multi-feature similarity function:
By the protein chain in seamless threading method comparison nonredundancy template base relative to search sequence each residue position
(i, j), wherein i is search sequence resi-dues to score f put, and j is fragment structure；F (i, j) in, under
Mark q represents search sequence feature score item, and subscript t represents template protein feature score item, Pq(i,k)
The sequence frequency obtained by PSI-BLAST for search sequence is composed, and wherein k is predetermined number amino acids
Type；Lq(i, k) and Lt(j k) is the search sequence obtained by PSI-BLAST and template sequence logarithmic spectrum；
sstJ () is the classification of template protein secondary structure, DSSP be calculated；ssqI () is search sequence two grades
Textural classification, can be obtained by two layers of neural metwork training；sat(j) and saqI () is formwork structure and search sequence
Solvent accessibility index, is trained by EDTSurf and neural network procedure and obtains；ψqJ () is inquiry
Sequence dihedral angle is to being obtained by two layers of neural metwork training；ψtJ () passes through query protein word
Allusion quotation obtains；SPt(j k) is the structure spectrum of template protein；w1、w2、w3、w4And w5For weighted value；
4) according to similarity score f, (i j) chooses M fragment of highest scoring on each position of search sequence and obtains fragment
Library file；
5) statistical query sequence is to the distance between coming from a template segments, and the most only statistics is less than's
Residue to spacing, draw rectangular histogram and obtain distance spectrum, the distance of rectangular histogram abscissa is spaced apart
When in template residue between distance in certain interval, then this interval sum just adds 1, if broken line graph existsThere is peak value in certain interior distance interval, then the distance interval that this peak value is corresponding is in target sequence
Residue i, to the Prediction distance of residue j, records the distance spectrum (profile) that this distribution is between two residues；
6) according to the position of residue in gained distance spectrum, search sequence is divided into n section；
7) l=1 is made, l ∈ 1,2,3 ..., and n}, operation following to segmental structure execution:
7.1) l section fragment structure is carried out fragment assembling；
7.2) calculate the distance between the resi-dues wherein containing distance spectrum information, and seek deviation with Prediction distance,
Cumulative deviation value is also averaged and is designated as Δ D；
7.3) if Δ D < R, then this conformation storage being designated as cell, wherein R is structural precision constraints；
7.4) 7.1 are repeated) to 7.3), until storage cell number reaches x cell, compare based on Rosetta Score 3
To the energy of segmental structure in cell, the structure choosing minimum energy is the pre-geodesic structure of this segmentation；
7.5) l=l+1, it is judged that whether l, more than or equal to n, is to enter 8), otherwise return to 7.1)；
8) arranging iterations is G, makes s=1, the following operation of execution:
8.1) target induced conformational ENERGY E (P is calculatedtarget)；
8.2) unsegmented structure is carried out fragment assembling, calculate energy value E (Ptrail)；
8.3) if E is (Ptarget)>E(Ptrail) then use PtrailReplace Ptarget；
8.4) s=s+1；Judge that s, whether more than or equal to G, is to enter 9), otherwise return to 8.1)；
9) output induced conformational, obtains search sequence nearly native state structure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610139514.4A CN105808972A (en) | 2016-03-11 | 2016-03-11 | Method for predicting protein structure from local to global on basis of knowledge spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610139514.4A CN105808972A (en) | 2016-03-11 | 2016-03-11 | Method for predicting protein structure from local to global on basis of knowledge spectrum |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105808972A true CN105808972A (en) | 2016-07-27 |
Family
ID=56467142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610139514.4A Pending CN105808972A (en) | 2016-03-11 | 2016-03-11 | Method for predicting protein structure from local to global on basis of knowledge spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105808972A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106372456A (en) * | 2016-08-26 | 2017-02-01 | 浙江工业大学 | Deep learning Residue2vec-based protein structure prediction method |
| CN107180164A (en) * | 2017-04-19 | 2017-09-19 | 浙江工业大学 | A kind of multiple domain protein structure assemble method based on template |
| CN107273713A (en) * | 2017-05-26 | 2017-10-20 | 浙江工业大学 | A kind of multiple domain protein template searching method based on TM align |
| CN107609340A (en) * | 2017-07-24 | 2018-01-19 | 浙江工业大学 | A kind of multiple domain albumen distance spectrum construction method |
| CN109360597A (en) * | 2018-08-01 | 2019-02-19 | 浙江工业大学 | A kind of group's Advances in protein structure prediction based on global and local policy cooperation |
| CN109378034A (en) * | 2018-08-28 | 2019-02-22 | 浙江工业大学 | A kind of protein prediction method based on range distribution estimation |
| CN112585685A (en) * | 2018-09-21 | 2021-03-30 | 渊慧科技有限公司 | Machine learning to determine protein structure |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205348A (en) * | 2015-09-22 | 2015-12-30 | 浙江工业大学 | Method for colony conformation space optimization based on distance constraint selection strategy |
-
2016
- 2016-03-11 CN CN201610139514.4A patent/CN105808972A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205348A (en) * | 2015-09-22 | 2015-12-30 | 浙江工业大学 | Method for colony conformation space optimization based on distance constraint selection strategy |
Non-Patent Citations (2)
| Title |
|---|
| DONG XU AND YANG ZHANG: "Ab initio protein structure assembly using continuous structure fragments and optimized konwledge-based force field", 《PROTEINS: STRUCTURE, FUNCTION, BIOINFORMATICS》 * |
| DONG XU AND YANG ZHANG: "Toward optimal fragment generations for ab initio protein structure assembly", 《PROTEINS: STRUCTURE, FUNCTION, BIOINFORMATICS》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106372456B (en) * | 2016-08-26 | 2019-01-22 | 浙江工业大学 | A kind of Advances in protein structure prediction based on deep learning |
| CN106372456A (en) * | 2016-08-26 | 2017-02-01 | 浙江工业大学 | Deep learning Residue2vec-based protein structure prediction method |
| CN107180164A (en) * | 2017-04-19 | 2017-09-19 | 浙江工业大学 | A kind of multiple domain protein structure assemble method based on template |
| CN107273713A (en) * | 2017-05-26 | 2017-10-20 | 浙江工业大学 | A kind of multiple domain protein template searching method based on TM align |
| CN107273713B (en) * | 2017-05-26 | 2020-06-02 | 浙江工业大学 | Multi-domain protein template searching method based on TM-align |
| CN107609340B (en) * | 2017-07-24 | 2020-05-05 | 浙江工业大学 | Multi-domain protein distance spectrum construction method |
| CN107609340A (en) * | 2017-07-24 | 2018-01-19 | 浙江工业大学 | A kind of multiple domain albumen distance spectrum construction method |
| CN109360597B (en) * | 2018-08-01 | 2020-12-01 | 浙江工业大学 | Group protein structure prediction method based on global and local strategy cooperation |
| CN109360597A (en) * | 2018-08-01 | 2019-02-19 | 浙江工业大学 | A kind of group's Advances in protein structure prediction based on global and local policy cooperation |
| CN109378034A (en) * | 2018-08-28 | 2019-02-22 | 浙江工业大学 | A kind of protein prediction method based on range distribution estimation |
| CN109378034B (en) * | 2018-08-28 | 2021-06-18 | 浙江工业大学 | Protein prediction method based on distance distribution estimation |
| CN112585685A (en) * | 2018-09-21 | 2021-03-30 | 渊慧科技有限公司 | Machine learning to determine protein structure |
| CN112585686A (en) * | 2018-09-21 | 2021-03-30 | 渊慧科技有限公司 | Machine learning to determine protein structure |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105808972A (en) | Method for predicting protein structure from local to global on basis of knowledge spectrum | |
| Li et al. | POSREG: proteomic signature discovered by simultaneously optimizing its reproducibility and generalizability | |
| Blaabjerg et al. | Rapid protein stability prediction using deep learning representations | |
| Ceriotti et al. | Demonstrating the transferability and the descriptive power of sketch-map | |
| Eckert et al. | Molecular similarity analysis in virtual screening: foundations, limitations and novel approaches | |
| Pierce et al. | A combination of rescoring and refinement significantly improves protein docking performance | |
| Viswanath et al. | Improving ranking of models for protein complexes with side chain modeling and atomic potentials | |
| Hou et al. | SeRenDIP-CE: sequence-based interface prediction for conformational epitopes | |
| Zheng et al. | Improving deep learning protein monomer and complex structure prediction using DeepMSA2 with huge metagenomics data | |
| Anishchenko et al. | Structural templates for comparative protein docking | |
| CA2942106A1 (en) | Aligning and clustering sequence patterns to reveal classificatory functionality of sequences | |
| Kmiecik et al. | Towards the high-resolution protein structure prediction. Fast refinement of reduced models with all-atom force field | |
| Zhang et al. | Predicting linear B-cell epitopes by using sequence-derived structural and physicochemical features | |
| Oldfield et al. | Computational prediction of secondary and supersecondary structures from protein sequences | |
| Dal Palu et al. | CLP-based protein fragment assembly | |
| CN105468934A (en) | Method for constructing distance model between protein residues based on Bolzmann probability density function | |
| Liu et al. | Assessing protein model quality based on deep graph coupled networks using protein language model | |
| Arshad et al. | An explainable stacking-based approach for accelerating the prediction of antidiabetic peptides | |
| Shi et al. | A review of machine learning-based methods for predicting drug–target interactions | |
| Hassan et al. | Classify a protein domain using sigmoid support vector machine | |
| US8024127B2 (en) | Local-global alignment for finding 3D similarities in protein structures | |
| Wang et al. | A sequence-based computational method for prediction of MoRFs | |
| Geethu et al. | Improved 3-D protein structure predictions using deep ResNet model | |
| Kern et al. | Predicting interacting residues using long-distance information and novel decoding in hidden markov models | |
| Hashmi et al. | Informatics-driven protein-protein docking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160727 |
|
| RJ01 | Rejection of invention patent application after publication |