Jpred
{{Context|date=May 2018}}
Jpred v.4 is the latest version of the JPred Protein Secondary Structure Prediction Server{{cite web|title=JPred4: A Protein Secondary Structure Prediction Server|url=http://www.compbio.dundee.ac.uk/jpred4/|accessdate=16 July 2015}} which provides predictions by the JNet algorithm, one of the most accurate methods for secondary structure prediction,{{cite journal|last1=Drozdetskiy|first1=Alexey|last2=Cole|first2=Chris|last3=Procter|first3=James|last4=Barton|first4=Geoffrey|title=JPred4: a protein secondary structure prediction server|journal=Nucleic Acids Research|date=Apr 16, 2015|doi=10.1093/nar/gkv332|volume=43|issue=W1|pages=W389–W394|pmid=25883141|pmc=4489285}} that has existed since 1998 in different versions.{{cite news | url=http://www.compbio.dundee.ac.uk/jpred4/old_news.shtml | title=JPred old news | date=Oct 25, 1998 | accessdate=16 Jul 2015}}
In addition to protein secondary structure, JPred also makes predictions of solvent accessibility and coiled-coil regions. The JPred service runs up to 134 000 jobs per month and has carried out over 2 million predictions in total for users in 179 countries.{{cite web|title=JPred4 statistics|url=http://www.compbio.dundee.ac.uk/jpred4/stats.shtml|accessdate=16 July 2015}}
JPred 2
The static HTML pages of JPred 2 are still available for reference.{{cite web|url=http://www.compbio.dundee.ac.uk/jpred4/legacy/|title=JPred2: legacy|accessdate=16 July 2015}}
JPred 3
The JPred v3{{cite web|url=http://www.compbio.dundee.ac.uk/jpred3/index.html|title=JPred3: previous version of JPred|accessdate=16 July 2015}} followed on from previous versions of JPred developed and maintained by James Cuff and Jonathan Barber (see JPred References{{cite web|title=JPred4 references|url=http://www.compbio.dundee.ac.uk/jpred/refs.shtml|accessdate=16 July 2015}}). This release added new functionality and fixed many bugs. The highlights are:
- New, friendlier user interface
- Retrained and optimised version of Jnet (v2) - mean secondary structure prediction accuracy of >81%
- Batch submission of jobs
- Better error checking of input sequences/alignments
- Predictions now (optionally) returned via e-mail
- Users may provide their own query names for each submission
- JPred now makes a prediction even when there are no PSI-BLAST hits to the query
- PS/PDF output now incorporates all the predictions
JPred 4
The current version of JPred (v4) has the following improvements and updates incorporated:
- Retrained on the latest UniRef90 and SCOPe/ASTRAL version of Jnet (v2.3.1) - mean secondary structure prediction accuracy of >82%.
- Upgraded the Web Server to the latest technologies (Bootstrap framework, JavaScript) and updating the web pages – improving the design and usability through implementing responsive technologies.
- Added RESTful API and mass-submission and results retrieval scripts - resulting in peak throughput above 20,000 predictions per day.{{cite web|title=JPred4 RESTful API|url=http://www.compbio.dundee.ac.uk/jpred4/api.shtml|accessdate=16 July 2015}}
- Added prediction jobs monitoring tools.{{cite web|title=JPred4 monitoring tools|url=http://www.compbio.dundee.ac.uk/jpred4/monitor.shtml|accessdate=16 July 2015}}
- Upgraded the results reporting – both, on the web-site, and through the optional email summary reports: improved batch submission, added results summary preview through Jalview results visualization summary in SVG and adding full multiple sequence alignments into the reports.
- Improved help-pages, incorporating tool-tips, and adding one-page step-by-step tutorials.{{cite web|title=JPred4 Help and Tutorials|url=http://www.compbio.dundee.ac.uk/jpred4/help.shtml|accessdate=16 July 2015}}
Sequence residues are categorised or assigned to one of the secondary structure elements, such as alpha-helix, beta-sheet and coiled-coil.
Jnet uses two neural networks for its prediction. The first network is fed with a window of 17 residues over each amino acid in the alignment plus a conservation number. It uses a hidden layer of nine nodes and has three output nodes, one for each secondary structure element.
The second network is fed with a window of 19 residues (the result of first network) plus the conservation number. It has a hidden layer with nine nodes and has three output nodes.{{cite journal | pmid = 10861942 | volume=40 | title=Application of multiple sequence alignment profiles to improve protein secondary structure prediction | date=August 2000 | journal=Proteins | pages=502–11 | last1 = Cuff | first1 = JA | last2 = Barton | first2 = GJ | issue=3 | doi=10.1002/1097-0134(20000815)40:3<502::aid-prot170>3.0.co;2-q| s2cid=855816 }}