Welcome to the CSEARCH / NMRPREDICT-Server

Lehre im Wintersemester 2014

The Austrian Way of University Funding

(seen on April 2^nd, 2014 in „Salzburger Nachrichten“)

Forschung und Leere …..

CSEARCH-Derived Web-Services
Peer-Reviewing of Carbon-NMR-Spectra Registration via email to Wolfgang.Robien[at]univie.ac.at is necessary before first usage Why do we need such a Robot-Referee for ¹³C-NMR assignments based on high-quality data and excellent algorithms ? The publication The referee-report The complete story I am quite sure you agree, that we do not want to solve future structure elucidation problems based on such "reference material" !	Web-Interface for the "CSEARCH-Robot-Referee" Web-Interface for the "CSEARCH-Robot-Referee" (Mirrored) (Both need Javascript) New release of the CSEARCH-Robot-Referee from 2014-12-06 (Documentation) New features: Better presentation of the result of the evaluation (feedback from the Berlin 2014 presentation) Summary of all evaluations performed for a particular compound can be recalled (see table at the end of each evaluation page) Every evaluation has a digital signature which can be verified online – manipulation of the resulting page is therefore impossible (the chance to manipulate a page successfully is below 1 : 10**40 ! For better understanding, this corresponds to the successful isolation of one molecule out of approximately 10 PetaMoles of substance) The underlying distribution of each predicted value is analyzed in detail – The result is given together with the predicted chemical shift value The Robot-referee evaluates a request as 'Major Revision' or 'Reject' – A similarity search using 61,000,000 predicted CNMR-spectra based on the PUBCHEM-Compounds is automatically launched
Using/Supporting the CSEARCH-Robot-Referee	European Journal of Organic Chemistry NMRSHIFTDB
Presentations of the CSEARCH-Robot-Referee	Erlangen 2010 Erlangen 2011 Erlangen 2012 Erlangen 2013 Köln 2013 Bochum 2013 Erlangen 2014 Darmstadt 2014 Mainz 2014 Berlin 2014 ( Dr. Liermann / University of Mainz ) Linz 2014 Erlangen 2015
Status of the CSEARCH-Robot-Referee	Information about the CSEARCH-Robot-Referee Information about scheduled maintenance
A few 'Highlights' of assignment errors as found in the literature	Assignment errors
Spectral similarity-search based on 16,4 millions predicted CNMR-spectra calculated for the PUBCHEM-structures including ranking of resulting hitlist	Spectral similarity search with ranking 16,4 millions of predicted CNMR-spectra
Spectral similarity-search based on 71 millions predicted CNMR-spectra calculated for the PUBCHEM-structures without ranking of the resulting hitlist	Spectral similarity search without ranking 71 millions of predicted CNMR-spectra / Pattern search
Spectral similarity-search based on 61 millions predicted CNMR-spectra calculated for the PUBCHEM-structures with ranking, separate deviation for every shiftvalue and unknown multiplicities possible	Spectral similarity search with ranking – latest version Test-version: Feedback highly appreciated 61 millions of predicted CNMR-spectra
Upgrade of CSEARCH-Databases including literature citations	Development-Version: CSEARCH-Upgrade Release-Version: Databases within NMRPREDICT / MestreNova
Searching for experimental C,P,F,N,O,B,Si-spectra available in a large variety of database- and prediction-programs	Search for experimental spectra via INCHIKEYS
Searching for existing experimental spectral data automatically from other programs	INCHIKEY-Description and automatic search
Cooperation partners using CSEARCH-data and/or CSEARCH-technology	CAS-Registryfile HaveItAll / KnowItAll MestReNova / NMRPredict&Mnova Verify NMRPredict NMRShiftDB SPECINFO TOPSPIN / CMC-se WILEY - European Journal of Organic Chemistry

CSEARCH-Derived Web-Services

Peer-Reviewing of Carbon-NMR-Spectra

Registration via email to Wolfgang.Robien[at]univie.ac.at

is necessary before first usage

Why do we need such a Robot-Referee for ¹³C-NMR assignments
based on high-quality data and excellent algorithms ?

The publication
The referee-report
The complete story

I am quite sure you agree, that we do not want to solve future
structure elucidation problems based on such "reference material" !

Web-Interface for the "CSEARCH-Robot-Referee"

Web-Interface for the "CSEARCH-Robot-Referee" (Mirrored)

(Both need Javascript)

New release of the CSEARCH-Robot-Referee from 2014-12-06 (Documentation)

New features:

Better presentation of the result of the evaluation (feedback from the Berlin 2014 presentation)
Summary of all evaluations performed for a particular compound can be recalled (see table at the end of each evaluation page)
Every evaluation has a digital signature which can be verified online – manipulation of the resulting page is therefore impossible (the chance to manipulate a page successfully is below 1 : 10**40 ! For better understanding, this corresponds to the successful isolation of one molecule out of approximately 10 PetaMoles of substance)
The underlying distribution of each predicted value is analyzed in detail – The result is given together with the predicted chemical shift value
The Robot-referee evaluates a request as 'Major Revision' or 'Reject' – A similarity search using 61,000,000 predicted CNMR-spectra based on the PUBCHEM-Compounds is automatically launched

Using/Supporting the CSEARCH-Robot-Referee

European Journal of Organic Chemistry

NMRSHIFTDB

Presentations of the CSEARCH-Robot-Referee

Erlangen 2010

Erlangen 2011

Erlangen 2012

Erlangen 2013

Köln 2013

Bochum 2013

Erlangen 2014

Darmstadt 2014

Mainz 2014

Berlin 2014 ( Dr. Liermann / University of Mainz )

Linz 2014

Erlangen 2015

Status of the CSEARCH-Robot-Referee

Information about the CSEARCH-Robot-Referee

Information about scheduled maintenance

A few 'Highlights' of assignment errors as found in the literature

Assignment errors

Spectral similarity-search based on 16,4 millions predicted CNMR-spectra calculated for the PUBCHEM-structures including ranking of resulting hitlist

16,4 millions of predicted CNMR-spectra

Spectral similarity-search based on 71 millions predicted CNMR-spectra calculated for the PUBCHEM-structures without ranking of the resulting hitlist

71 millions of predicted CNMR-spectra / Pattern search

Spectral similarity-search based on 61 millions predicted CNMR-spectra calculated for the PUBCHEM-structures with ranking, separate deviation for every shiftvalue and unknown multiplicities possible

Test-version: Feedback highly appreciated

61 millions of predicted CNMR-spectra

Upgrade of CSEARCH-Databases including literature citations

Development-Version: CSEARCH-Upgrade

Release-Version: Databases within NMRPREDICT / MestreNova

Searching for experimental C,P,F,N,O,B,Si-spectra available in a large variety of database- and prediction-programs

Search for experimental spectra via INCHIKEYS

Searching for existing experimental spectral data automatically from other programs

INCHIKEY-Description and automatic search

Cooperation partners using CSEARCH-data and/or CSEARCH-technology

CAS-Registryfile

HaveItAll / KnowItAll