Published online 2 November 2015 Nucleic Acids Research, 2016, Vol. 44, Database issue D1113–D1118 doi: 10.1093/nar/gkv1143 Norine, the knowledgebase dedicated to non-ribosomal peptides, is now open to crowdsourcing Areski Flissi 1,2,* , Yoann Dufresne 1,2 , Juraj Michalik 1,3 , Laurie Tonon 1,2 , St ´ ephane Janot 1,2 , Laurent No ´ e 1,2 , Philippe Jacques 4 , Val´ erie Lecl ` ere 1,2,4 and Maude Pupin 1,2,3,* 1 University of Lille, CRIStAL, UMR CNRS 9189, cit ´ e scientifique––bat M3ext, 59650 Villeneuve d’Ascq, France, 2 Inria Lille Nord Europe, Bonsai team, Parc scientifique de la Haute Borne, 40, avenue Halley––Bt A, 59650 Villeneuve d’Ascq, France, 3 University of Lille, bilille, cit´ e scientifique––bat M3ext, 59650 Villeneuve d’Ascq, France and 4 University of Lille, EA 7394, ICV-Institut Charles Viollette, ProBioGEM team, Polytech’Lille, avenue Langevin, 59655 Villeneuve d’Ascq, France Received September 15, 2015; Revised October 14, 2015; Accepted October 16, 2015 ABSTRACT Since its creation in 2006, Norine remains the unique knowledgebase dedicated to non-ribosomal pep- tides (NRPs). These secondary metabolites, pro- duced by bacteria and fungi, harbor diverse inter- esting biological activities (such as antibiotic, anti- tumor, siderophore or surfactant) directly related to the diversity of their structures. The Norine team goal is to collect the NRPs and provide tools to analyze them efficiently. We have developed a user-friendly interface and dedicated tools to provide a complete bioinformatics platform. The knowledgebase gathers abundant and valuable annotations on more than 1100 NRPs. To increase the quantity of described NRPs and improve the quality of associated anno- tations, we are now opening Norine to crowdsourc- ing. We believe that contributors from the scientific community are the best experts to annotate the NRPs they work on. We have developed MyNorine to facil- itate the submission of new NRPs or modifications of stored ones. This article presents MyNorine and other novelties of Norine interface released since the first publication. Norine is freely accessible from the following URL: http://bioinfo.lifl.fr/NRP. INTRODUCTION Norine, first released in 2006 (1), remains the unique plat- form dedicated to computational biology analysis of non- ribosomal peptides (NRPs). Although other databases con- tain NRPs, they did not have the same scope and did not provide associated analysis tools. For example, the pep- taibiotics database (2) is a comprehensive online resource for peptaibols, while ClusterMine360 (3) is a database of microbial PKS/NRPS Biosynthesis. The NRPs have in- creased in popularity in recent years because they har- bor diverse interesting biological activities. Indeed, they are produced by micro-organisms, bacteria and fungi, to col- onize and survive in various environments. Among oth- ers, NRPs can act as antibiotics (penicillin -NOR00006- , daptomycin -NOR00001- or vancomycin -NOR00681-), siderophores (pyoverdins -NOR00160 to 206, NOR00903 to 912- or vibriobactin -NOR00250-), surfactants or pro- tease inhibitors. In addition to their primary activity, some NRPs are also successfully prescribed for treating cancers (actinomycin D -NOR00228-) or reducing transplant rejec- tion (cyclosporin A -NOR00033-). Beyond the pharmacol- ogy, NRPs promise other advantageous applications such as biocontrol of plant diseases, bioremediation of areas con- taminated with toxic metals and/or non-biodegradable or- ganic compounds. These metabolites are produced by a specific biosynthetic pathway. In few words, huge enzymes called NRP synthetases select specific amino acids and as- semble them by peptide bonds. Auxiliary enzymes mod- ify the produced peptides by adding other types of build- ing blocks (monomers) and/or by forming other types of bonds. The diversity of NRPs resides in their uncommon biosynthetic pathways (for recent reviews see (4,5)). As it gathers more than 1100 NRPs, Norine enables to explore and better understand the diversity of the NRPs. We already suggested that their diversity of activities comes from the diversity of their structures (6). The most partic- ular and interesting feature is the diversity of composition. Norine hosts 529 different monomers including proteino- genic amino acids and their derivatives, other amino acids, lipids, carbohydrates, chromophores and polyketides. An- * To whom correspondence should be addressed. Tel: +33 328 77 85 60; Fax: +33 328 77 85 37; Email: areski.fl[email protected]Correspondence may also be addressed to Maude Pupin. Tel: +33 328 77 85 55; Fax: +33 328 77 85 37; Email: [email protected]C The Author(s) 2015. Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected]Downloaded from https://academic.oup.com/nar/article-abstract/44/D1/D1113/2502639 by guest on 15 February 2018
Transcript
Published online 2 November 2015 Nucleic Acids Research 2016 Vol 44 Database issue D1113ndashD1118doi 101093nargkv1143
Norine the knowledgebase dedicated tonon-ribosomal peptides is now open tocrowdsourcingAreski Flissi12 Yoann Dufresne12 Juraj Michalik13 Laurie Tonon12 Stephane Janot12Laurent Noe12 Philippe Jacques4 Valerie Leclere124 and Maude Pupin123
1University of Lille CRIStAL UMR CNRS 9189 cite scientifiquendashndashbat M3ext 59650 Villeneuve drsquoAscq France 2InriaLille Nord Europe Bonsai team Parc scientifique de la Haute Borne 40 avenue HalleyndashndashBt A 59650 VilleneuvedrsquoAscq France 3University of Lille bilille cite scientifiquendashndashbat M3ext 59650 Villeneuve drsquoAscq France and4University of Lille EA 7394 ICV-Institut Charles Viollette ProBioGEM team PolytechrsquoLille avenue Langevin 59655Villeneuve drsquoAscq France
Received September 15 2015 Revised October 14 2015 Accepted October 16 2015
ABSTRACT
Since its creation in 2006 Norine remains the uniqueknowledgebase dedicated to non-ribosomal pep-tides (NRPs) These secondary metabolites pro-duced by bacteria and fungi harbor diverse inter-esting biological activities (such as antibiotic anti-tumor siderophore or surfactant) directly related tothe diversity of their structures The Norine team goalis to collect the NRPs and provide tools to analyzethem efficiently We have developed a user-friendlyinterface and dedicated tools to provide a completebioinformatics platform The knowledgebase gathersabundant and valuable annotations on more than1100 NRPs To increase the quantity of describedNRPs and improve the quality of associated anno-tations we are now opening Norine to crowdsourc-ing We believe that contributors from the scientificcommunity are the best experts to annotate the NRPsthey work on We have developed MyNorine to facil-itate the submission of new NRPs or modificationsof stored ones This article presents MyNorine andother novelties of Norine interface released since thefirst publication Norine is freely accessible from thefollowing URL httpbioinfoliflfrNRP
INTRODUCTION
Norine first released in 2006 (1) remains the unique plat-form dedicated to computational biology analysis of non-ribosomal peptides (NRPs) Although other databases con-tain NRPs they did not have the same scope and did notprovide associated analysis tools For example the pep-
taibiotics database (2) is a comprehensive online resourcefor peptaibols while ClusterMine360 (3) is a database ofmicrobial PKSNRPS Biosynthesis The NRPs have in-creased in popularity in recent years because they har-bor diverse interesting biological activities Indeed they areproduced by micro-organisms bacteria and fungi to col-onize and survive in various environments Among oth-ers NRPs can act as antibiotics (penicillin -NOR00006- daptomycin -NOR00001- or vancomycin -NOR00681-)siderophores (pyoverdins -NOR00160 to 206 NOR00903to 912- or vibriobactin -NOR00250-) surfactants or pro-tease inhibitors In addition to their primary activity someNRPs are also successfully prescribed for treating cancers(actinomycin D -NOR00228-) or reducing transplant rejec-tion (cyclosporin A -NOR00033-) Beyond the pharmacol-ogy NRPs promise other advantageous applications suchas biocontrol of plant diseases bioremediation of areas con-taminated with toxic metals andor non-biodegradable or-ganic compounds These metabolites are produced by aspecific biosynthetic pathway In few words huge enzymescalled NRP synthetases select specific amino acids and as-semble them by peptide bonds Auxiliary enzymes mod-ify the produced peptides by adding other types of build-ing blocks (monomers) andor by forming other types ofbonds The diversity of NRPs resides in their uncommonbiosynthetic pathways (for recent reviews see (45))
As it gathers more than 1100 NRPs Norine enables toexplore and better understand the diversity of the NRPsWe already suggested that their diversity of activities comesfrom the diversity of their structures (6) The most partic-ular and interesting feature is the diversity of compositionNorine hosts 529 different monomers including proteino-genic amino acids and their derivatives other amino acidslipids carbohydrates chromophores and polyketides An-
To whom correspondence should be addressed Tel +33 328 77 85 60 Fax +33 328 77 85 37 Email areskiflissiuniv-lille1frCorrespondence may also be addressed to Maude Pupin Tel +33 328 77 85 55 Fax +33 328 77 85 37 Email maudepupinuniv-lille1fr
Ccopy The Author(s) 2015 Published by Oxford University Press on behalf of Nucleic Acids ResearchThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby-nc40) whichpermits non-commercial re-use distribution and reproduction in any medium provided the original work is properly cited For commercial re-use please contactjournalspermissionsoupcom
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1114 Nucleic Acids Research 2016 Vol 44 Database issue
other feature is the diversity of structures Not all NRPsare linear they can also contain one or more cycles andbranches Therefore we provide a specific representation ofNRPs monomeric structures that we call the NORINE for-mat The monomeric structures and their associated home-made bioinformatics tools are major features of Norine
Norine is freely accessible from the following URL httpbioinfoliflfrnorine It gives access to abundant and valu-able annotations on NRPs through a user-friendly interfaceThe data can be browsed searched (among annotations andstructures) downloaded and accessed by web APIs More-over the peptides and their annotations are manually ex-tracted from scientific literature So expert and meticulouswork is needed to populate the knowledgebase The highdata quality has contributed to the renown of Norine Sci-entists worldwide rely on Norine for studying and discov-ering new natural products by for example combining ge-nomics and metabolomics (78) Some mass spectra toolsintegrate Norine data as mMass (9) or iSNAP (10) More-over the Worldwide Protein DataBank (wwPBD) organi-zation selected Norine as a reference database (11) Nowwe are opening Norine to crowdsourcing Indeed we be-lieve that contributors from the scientific community arethe best experts to annotate the NRPs they work on Wedeveloped MyNorine that eases the process of entering andsubmitting a new NRP or a modification of a stored oneThe submissions are then manually accepted by validatorsof the Norine team All partners benefit from crowdsourc-ing Norine knowledgebase is improved and contributorsreceive recognition by being cited as authors of the Norineentries they fill in Their publications can also be entered inNorine another way to increase the authors visibility
This article describes in details MyNorine and the asso-ciated submission process It also presents the novelties de-veloped since the first publication of Norine in 2008
CROWDSOURCING FOR INCREASING DATA QUAN-TITY AND QUALITY MYNORINE TOOL
MyNorine is a new module that is now part of the Norineplatform With this tool the main idea is to enhance thefirst database entirely dedicated to NRPs by allowing sci-entists all over the world to contribute From users pointof view MyNorine can be seen as an interface to simplifythe submission of new peptides into the Norine databaseand to improve quality of information stored by submit-ting proposals to correct or modify existing peptides Thetool is mainly dedicated to biologists and biochemists whowant to contribute to the Norine database In order touse MyNorine users firstly register by creating an accountMyNorine distinguishes two main roles curators and val-idators Curators contribute to the Norine database by sub-mitting new peptides or modifications whereas validatorsare responsible for validation steps Access to the differ-ent interfaces and actions in MyNorine depends on usersrsquorights which are determined by their role
Overview
Figure 1 gives an overview of how MyNorine worksthrough the main use cases For curators the two main fea-tures are (i) submission of new NRPs and (ii) proposal
of modification of an existing entry Regarding validatorswhich are responsible for the final acceptation of submis-sions or modifications MyNorine provides specific inter-faces and dashboards
Submission of NRPs
Submissions of NRPs can be achieved by different man-ners (i) successively fill in different forms correspondingto different classes of annotations and finally send a noti-fication to validators (ii) use either XML or JSON (httpwwwjsonorg) files that represent a NRP (generated byother tools or duplicated from an existing entry of Norine)
Submit a new peptide by completing forms MyNorine pro-vides an interface for submitting proposals of new NRPsThe interface is composed of several forms that roughlycorrespond to the different classes of annotations Theseformsclasses are
(i) general information such as the name the familythe known activities the empirical chemical formula themolecular weight etc
(ii) structure features type (cyclic linear) monomericcomposition monomeric structure 2D chemical graph andSMILES if available
(iii) producing organisms and their taxonomy(iv) published references and authors associated to the
peptide(v) links to other resources such as PDB UniProt and
PubChem with the associated accession numbers(vi) any complementary general informationTo help and guide the curators MyNorine offers so-
phisticated interfaces in order to interact with them dur-ing the completion of the forms For example it will au-tomatically suggest peptide names monomer codes refer-ences or authors (more generally all information stored inthe database when it matches the user entry) Furthermoreusers are warned if a peptide already exists Submission ofa new peptide by curators triggers a workflow process inwhich validators are notified The new peptide with all as-sociated annotations is manually checked eventually mod-ified or corrected before the acceptation of the proposalFinally in case of approval the curator is notified and willbe mentioned as the author of the entry in the page of thepeptide entry
Alternative ways for submitting peptides In parallel tomanual input for new NRPs MyNorine supports XML orJSON files as an alternative way to submit data We de-fined a simple representation for NRPs annotations usingXMLJSON notations The choice of these notations andespecially the JSON data-interchange format was leaded byits ability to be both comprehensive by humans and easilymanipulated by computer programs independently of thelanguage XMLJSON files can be automatically generatedfrom external sources (other databases) and directly submit-ted to Norine Otherwise a curator can upload an XML orJSON file that describes partially or completely a NRP toMyNorine Once uploaded forms are automatically com-pleted thanks to data extracted from the file Benefits aremanifold Users do not need to complete the whole process
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1115
Figure 1 Overview of MyNorine the main use cases
at once Indeed users can at any time during the processdownload XML or JSON file of their current productionthen work locally on it (or not) and finally upload the file toMyNorine later In other words downloading a NRP fileenables to work off-line The submission process continuesnormally after uploading Furthermore it is possible to usean existing peptide as a template for instance to input a newvariant of a peptide already in Norine To use this featurecurators duplicate the NRP from its result page as illus-trated in Figure 2
Proposal of modification for existing entries
To improve quality of information stored in the Norinedatabase and ensure its integrity MyNorine provides a pro-cess to suggest modifications for an entry Any annotationcan be corrected or added through an icon on the peptideresult page (see Figure 2) Users can load an entry add orchange one or more data and send a notification to the val-idators with a message pointing out what has been proposedto be modified After manual check validators accept or re-ject the proposal Accepted modifications are immediatelyvisible on peptide result pages In this case the curator ismentioned as the author of the modification of the peptideentry
Validation
Manual validation Regarding validation aspectMyNorine offers to validators specific interfaces andtwo dashboards (one for new peptides the other formodifications) as illustrated in Figure 3 for new peptidesThe dashboard is divided into three tabs The first tab listscurrent unprocessed submissions and allows validators toload check accept (or reject) them The second tab listssubmissions that are under validation process The last one
shows accepted submissions and their history (submissiondate contributor)
Validators ensure that NRPs hosted in Norine are man-ually extracted from scientific knowledge (literature otherdatabases personal analysis) The NRPs that are predictedexclusively by bioinformatics tools are avoided At least oneof the following evidences is required for a submission toNorine
(i) experimental elucidation of the complete structureeventually coupled with bioinformatics analysis of the cor-responding NRPS gene cluster or
(ii) experimental mass determination coupled with thecareful bioinformatics analysis of the corresponding NRPSgene cluster or
(iii) manually curated annotation obtained by similaritywith gene clusters coding for NRP synthetases studied ex-perimentally
All the peptides submitted through MyNorine will be in-spected by validators of the Norine team Only the peptidesmatching Norine standards will be validated and incorpo-rated into the public version of the database If no evidencefor the non-ribosomal origin is presented the validatorsmay assign the putative status to the submitted peptide
Semi-automatic verifications s2m tool Manual valida-tion is strengthened by semi-automatic verifications Ba-sic checks are done automatically to ensure the consis-tency of the data within and among the entries Above allthe monomeric structures are checked through s2m tools2m infers automatically monomeric structures from chem-ical structures represented by SMILES notation Chemi-cal structures of NRPs stored in Norine are extracted fromcompound databases such as PubChem (12) or PDB (13)The predicted monomeric structures are compared to theNorine structures We detected 97 incongruities between thestructures in Norine and the ones of other databases A re-
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1116 Nucleic Acids Research 2016 Vol 44 Database issue
Figure 2 Duplicate an existing NRP or propose modification
Figure 3 Dashboard for validators
mediation process started to identify from which databasescomes each error and to correct the corresponding annota-tions
FACILITATION OF NORINE DATA ACCESS
Since the creation of Norine the interface has been im-proved to ease the usage of the website the search for rele-vant data and their visualization The major improvementssince the 2008 publication are presented in the followingparagraphs
Improved Norine user interface
Access to monomers annotations A query form enablesto search for the 529 monomers constituting the NRPsstored in Norine They are clustered by class of compounds(carbohydrates fatty acids chromophores polyketides andtypes of amino acids) For each monomer chemical detailsare provided such as short name IUPAC name empiri-cal molecular formula molecular weight chemical struc-ture represented by two-dimensions chemical graphs andSMILES notation and links to PubChem (12) and theChemical Component Dictionary (14)
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1117
Efficient structure search The structure search query formhas been simplified The editor helping to draw monomericstructurecomposition of NRPs is now coded in JavaScriptThe interactivity with the user is improved with fast andsimple access to the monomer list and intuitive drawingfunctionalities A new algorithm was introduced (15) tosearch for a monomer list among the Norine peptides basedon a monomeric composition fingerprint (MCFP) Finallythe different algorithms (MCFP search pattern search andsimilarity search) were combined to output only one resultpage The peptides similar to the query are sorted by dis-tances For each found peptide a specific page shows thecomparison between it and the query This enhanced struc-ture search functionality is already operated by the anti-SMASH platform (16ndash18) When a NRPS gene cluster ispredicted a putative peptide is constructed and can be com-pared to the reference NRPs through a direct link to Norine
Filtering by graphical output The annotation or structuresearches output a result page that lists the found peptidesAn icon representing a pie chart gives access to a graphicalrepresentation Pie charts illustrate global numerical pro-portion of found peptides for several annotations The slicesare clickable and give access to the current peptide list lim-ited to the selected criteria This process allows users to filterthe results by more selective criteria
Download and customize the data Norine offers the op-portunity to download all data in several formats From apeptide or a monomer page icons give all annotations inTXTCSV HTML XML or JSON formats as illustratedin Figure 2 From the result pages listing a set of peptidesthe user can customize the annotations that will be inte-grated in the output With the HTML format a table withall the annotations selected is provided So the user cancompare the desired annotations of several peptides at atime
Norine REST web services
A web API to access the Norine database We developed aweb API (Application Programming Interface) to access theNorine database This API is based on REST (Represen-tational state transfer) REST (19) is an architectural stylebased on web-standards and the HTTP protocol Web APIsexpose resources to computer programs Whereas classicalweb approaches are mainly dedicated to humans by provid-ing rich and interactive interfaces (such as web forms or piediagrams) for querying the database Norine RESTful webservices enable integration with other resources by givingaccess to data it contains using simple resource URI Inmost of the cases data returned by such services are avail-able in various formats such as HTML XML or JSONCurrently Norine provides the following services
(i) retrieve NRP annotations from Norine ID or by name(ii) get list of all monomers with cluster tree(iii) get the access code of an external link or retrieve
database name (PDB PubChem) from access code(iv) obtain information on producing organisms by nameThe Norine REST services URIs
use the following general syntaxhttpnorinerestltpathgtltformatgt[parameter]
The path argument determines the type of service to useformat can be XML or JSON and parameter correspondsto the user query For example to retrieve annotations forthe vancomycin NRP in JSON format the URI will looklike httpnorinerestjsonnamevancomycin
Write programs for Norine REST services Another fea-ture that may interest developers of other resources anddatabases is the ability to access Norine through programsDevelopers can write client programs in any languageThese programs simply consist in creating HTTP requestswith the specific URI described above and handling the re-sponse It is also possible to embed its own HTML form toquery Norine and receive response in XML or JSON for-mats
CONCLUSION
Norine (httpbioinfoliflfrNRP) is the unique platformdedicated to NRPs A user-friendly interface eases thebrowsing annotation or structure searching and download-ing of the NRPs and their monomers This interface is com-pleted by programmatic access through a web API basedon REST Dedicated bioinformatics tools are associatedwith the database such as an editor and a visualizer of themonomeric structures homemade algorithms to comparemonomeric structures and s2m that infers monomeric struc-ture from chemical structure An important improvementis the development of MyNorine tool This tool is an in-terface to simplify the entry of new peptides or modifica-tion of existing ones in Norine by biologists The processintegrates the different steps going from the completion ofthe annotations to the validation of the submitted data Sci-entists can create an account and start entering informa-tion in MyNorine The submission will first be sent to thevalidators of Norine team before being made available inNorine Thanks to MyNorine the scientific community caneasily contribute to increase the quantity of identified NRPsand improve the quality of associated annotations storedin Norine Our goal is to initiate the crowdsourcing by ap-pealing for the authors of articles describing NRPs not inNorine to contribute to this resource The external contrib-utors will be promoted on Norine homepage and will beassociated to the NRP entries they fill in
ACKNOWLEDGEMENT
The authors would like to thank Mohcen Benmounah andAntoine Engelaere for their participation in Norine devel-opments
FUNDING
Bilille the bioinformatics service platform of Lille Univer-sity of Lille and Inria Funding for open access charge In-riaConflict of interest statement None declared
REFERENCES1 CabocheS PupinM LeclereV FontaineA JacquesP and
KucherovG (2008) NORINE a database of nonribosomal peptidesNucleic Acids Res 36 D326ndashD331
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1118 Nucleic Acids Research 2016 Vol 44 Database issue
2 NeumannNKN StoppacherN ZeilingerS DegenkolbTBrucknerH and SchuhmacherR (2015) The PeptaibioticsDatabasendashndasha comprehensive online resource Chem Biodivers 12743ndash751
3 ConwayKR and BoddyCN (2013) ClusterMine360 a database ofmicrobial PKSNRPS biosynthesis Nucleic Acids Res 41D402ndashD407
4 WalshCT (2016) Insights into the chemical logic and enzymaticmachinery of NRPS assembly lines Nat ProdRep DOI101039C5NP00035A
5 MarahielMA (2016) A structural model for multimodular NRPSassembly lines Nat Prod Rep DOI101039C5NP00082C
6 CabocheS LeclereV PupinM KucherovG and JacquesP (2010)Diversity of monomers in nonribosomal peptides towards theprediction of origin and biological activity J Bacteriol 1925143ndash5150
7 DoroghaziJR AlbrightJC GoeringAW JuK-S HainesRRTchalukovKA LabedaDP KelleherNL and MetcalfWW(2014) A roadmap for natural product discovery based on large-scalegenomics and metabolomics Nat Chem Biol 10 963ndash968
8 MedemaMH PaalvastY NguyenDD MelnikADorresteinPC TakanoE and BreitlingR (2014) Pep2Pathautomated mass spectrometry-guided genome mining of peptidicnatural products PLoS Comput Biol 10 e1003822
9 NiedermeyerTHJ and StrohalmM (2012) mMass as a softwaretool for the annotation of cyclic peptide tandem mass spectra PLoSOne 7 e44913
10 IbrahimA YangL JohnstonC LiuX MaB andMagarveyNA (2012) Dereplicating nonribosomal peptides using aninformatic search algorithm for natural products (iSNAP) discoveryProc Natl Acad Sci USA 109 19196ndash19201
11 VelankarS DanaJM JacobsenJ GinkelGv GanePJ LuoJOldfieldTJ OrsquoDonovanC MartinM-J andKleywegtGJ (2012) SIFTS structure integration with function
taxonomy and sequences resource Nucleic AcidsRes 41 D483ndashD489
12 BoltonEE WangY ThiessenPA andBryantSH (2008) PubChem Integrated Platform of SmallMolecules and Biological Activities Annu Rep ComputChem 4 217ndash241
13 BermanH HenrickK and NakamuraH (2003) Announcing theworldwide Protein Data Bank Nat Struct Mol Biol 10 980ndash980
14 WestbrookJD ShaoC FengZ ZhuravlevaM ValenkarS andYoungJ (2014) The chemical component dictionary completedescriptions of constituent molecules in experimentally determined3D macromolecules in the Protein DataBank Bioinformatics 31 1274ndash1278
15 AbdoA CabocheS LeclereV JacquesP and PupinM (2012) Anew fingerprint to predict nonribosomal peptides activity J ComputAided Mol Des 26 1187ndash1194
16 WeberT BlinK DuddelaS KrugD KimHU BruccoleriRLeeSY FischbachMA MullerR WohllebenW et al (2015)antiSMASH 30ndashndasha comprehensive resource for the genome miningof biosynthetic gene clusters Nucleic Acids Res 43 W237ndashW243
17 BlinK MedemaMH KazempourD FischbachMABreitlingR TakanoE and WeberT (2013) antiSMASH 20ndashndashaversatile platform for genome mining of secondary metaboliteproducers Nucleic Acids Res 41 W204ndashW212
19 FieldingRT and TaylorRN (2002) Principled design of themodern Web architecture ACM Trans Internet Technol 2 115ndash150
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1114 Nucleic Acids Research 2016 Vol 44 Database issue
other feature is the diversity of structures Not all NRPsare linear they can also contain one or more cycles andbranches Therefore we provide a specific representation ofNRPs monomeric structures that we call the NORINE for-mat The monomeric structures and their associated home-made bioinformatics tools are major features of Norine
Norine is freely accessible from the following URL httpbioinfoliflfrnorine It gives access to abundant and valu-able annotations on NRPs through a user-friendly interfaceThe data can be browsed searched (among annotations andstructures) downloaded and accessed by web APIs More-over the peptides and their annotations are manually ex-tracted from scientific literature So expert and meticulouswork is needed to populate the knowledgebase The highdata quality has contributed to the renown of Norine Sci-entists worldwide rely on Norine for studying and discov-ering new natural products by for example combining ge-nomics and metabolomics (78) Some mass spectra toolsintegrate Norine data as mMass (9) or iSNAP (10) More-over the Worldwide Protein DataBank (wwPBD) organi-zation selected Norine as a reference database (11) Nowwe are opening Norine to crowdsourcing Indeed we be-lieve that contributors from the scientific community arethe best experts to annotate the NRPs they work on Wedeveloped MyNorine that eases the process of entering andsubmitting a new NRP or a modification of a stored oneThe submissions are then manually accepted by validatorsof the Norine team All partners benefit from crowdsourc-ing Norine knowledgebase is improved and contributorsreceive recognition by being cited as authors of the Norineentries they fill in Their publications can also be entered inNorine another way to increase the authors visibility
This article describes in details MyNorine and the asso-ciated submission process It also presents the novelties de-veloped since the first publication of Norine in 2008
CROWDSOURCING FOR INCREASING DATA QUAN-TITY AND QUALITY MYNORINE TOOL
MyNorine is a new module that is now part of the Norineplatform With this tool the main idea is to enhance thefirst database entirely dedicated to NRPs by allowing sci-entists all over the world to contribute From users pointof view MyNorine can be seen as an interface to simplifythe submission of new peptides into the Norine databaseand to improve quality of information stored by submit-ting proposals to correct or modify existing peptides Thetool is mainly dedicated to biologists and biochemists whowant to contribute to the Norine database In order touse MyNorine users firstly register by creating an accountMyNorine distinguishes two main roles curators and val-idators Curators contribute to the Norine database by sub-mitting new peptides or modifications whereas validatorsare responsible for validation steps Access to the differ-ent interfaces and actions in MyNorine depends on usersrsquorights which are determined by their role
Overview
Figure 1 gives an overview of how MyNorine worksthrough the main use cases For curators the two main fea-tures are (i) submission of new NRPs and (ii) proposal
of modification of an existing entry Regarding validatorswhich are responsible for the final acceptation of submis-sions or modifications MyNorine provides specific inter-faces and dashboards
Submission of NRPs
Submissions of NRPs can be achieved by different man-ners (i) successively fill in different forms correspondingto different classes of annotations and finally send a noti-fication to validators (ii) use either XML or JSON (httpwwwjsonorg) files that represent a NRP (generated byother tools or duplicated from an existing entry of Norine)
Submit a new peptide by completing forms MyNorine pro-vides an interface for submitting proposals of new NRPsThe interface is composed of several forms that roughlycorrespond to the different classes of annotations Theseformsclasses are
(i) general information such as the name the familythe known activities the empirical chemical formula themolecular weight etc
(ii) structure features type (cyclic linear) monomericcomposition monomeric structure 2D chemical graph andSMILES if available
(iii) producing organisms and their taxonomy(iv) published references and authors associated to the
peptide(v) links to other resources such as PDB UniProt and
PubChem with the associated accession numbers(vi) any complementary general informationTo help and guide the curators MyNorine offers so-
phisticated interfaces in order to interact with them dur-ing the completion of the forms For example it will au-tomatically suggest peptide names monomer codes refer-ences or authors (more generally all information stored inthe database when it matches the user entry) Furthermoreusers are warned if a peptide already exists Submission ofa new peptide by curators triggers a workflow process inwhich validators are notified The new peptide with all as-sociated annotations is manually checked eventually mod-ified or corrected before the acceptation of the proposalFinally in case of approval the curator is notified and willbe mentioned as the author of the entry in the page of thepeptide entry
Alternative ways for submitting peptides In parallel tomanual input for new NRPs MyNorine supports XML orJSON files as an alternative way to submit data We de-fined a simple representation for NRPs annotations usingXMLJSON notations The choice of these notations andespecially the JSON data-interchange format was leaded byits ability to be both comprehensive by humans and easilymanipulated by computer programs independently of thelanguage XMLJSON files can be automatically generatedfrom external sources (other databases) and directly submit-ted to Norine Otherwise a curator can upload an XML orJSON file that describes partially or completely a NRP toMyNorine Once uploaded forms are automatically com-pleted thanks to data extracted from the file Benefits aremanifold Users do not need to complete the whole process
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1115
Figure 1 Overview of MyNorine the main use cases
at once Indeed users can at any time during the processdownload XML or JSON file of their current productionthen work locally on it (or not) and finally upload the file toMyNorine later In other words downloading a NRP fileenables to work off-line The submission process continuesnormally after uploading Furthermore it is possible to usean existing peptide as a template for instance to input a newvariant of a peptide already in Norine To use this featurecurators duplicate the NRP from its result page as illus-trated in Figure 2
Proposal of modification for existing entries
To improve quality of information stored in the Norinedatabase and ensure its integrity MyNorine provides a pro-cess to suggest modifications for an entry Any annotationcan be corrected or added through an icon on the peptideresult page (see Figure 2) Users can load an entry add orchange one or more data and send a notification to the val-idators with a message pointing out what has been proposedto be modified After manual check validators accept or re-ject the proposal Accepted modifications are immediatelyvisible on peptide result pages In this case the curator ismentioned as the author of the modification of the peptideentry
Validation
Manual validation Regarding validation aspectMyNorine offers to validators specific interfaces andtwo dashboards (one for new peptides the other formodifications) as illustrated in Figure 3 for new peptidesThe dashboard is divided into three tabs The first tab listscurrent unprocessed submissions and allows validators toload check accept (or reject) them The second tab listssubmissions that are under validation process The last one
shows accepted submissions and their history (submissiondate contributor)
Validators ensure that NRPs hosted in Norine are man-ually extracted from scientific knowledge (literature otherdatabases personal analysis) The NRPs that are predictedexclusively by bioinformatics tools are avoided At least oneof the following evidences is required for a submission toNorine
(i) experimental elucidation of the complete structureeventually coupled with bioinformatics analysis of the cor-responding NRPS gene cluster or
(ii) experimental mass determination coupled with thecareful bioinformatics analysis of the corresponding NRPSgene cluster or
(iii) manually curated annotation obtained by similaritywith gene clusters coding for NRP synthetases studied ex-perimentally
All the peptides submitted through MyNorine will be in-spected by validators of the Norine team Only the peptidesmatching Norine standards will be validated and incorpo-rated into the public version of the database If no evidencefor the non-ribosomal origin is presented the validatorsmay assign the putative status to the submitted peptide
Semi-automatic verifications s2m tool Manual valida-tion is strengthened by semi-automatic verifications Ba-sic checks are done automatically to ensure the consis-tency of the data within and among the entries Above allthe monomeric structures are checked through s2m tools2m infers automatically monomeric structures from chem-ical structures represented by SMILES notation Chemi-cal structures of NRPs stored in Norine are extracted fromcompound databases such as PubChem (12) or PDB (13)The predicted monomeric structures are compared to theNorine structures We detected 97 incongruities between thestructures in Norine and the ones of other databases A re-
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1116 Nucleic Acids Research 2016 Vol 44 Database issue
Figure 2 Duplicate an existing NRP or propose modification
Figure 3 Dashboard for validators
mediation process started to identify from which databasescomes each error and to correct the corresponding annota-tions
FACILITATION OF NORINE DATA ACCESS
Since the creation of Norine the interface has been im-proved to ease the usage of the website the search for rele-vant data and their visualization The major improvementssince the 2008 publication are presented in the followingparagraphs
Improved Norine user interface
Access to monomers annotations A query form enablesto search for the 529 monomers constituting the NRPsstored in Norine They are clustered by class of compounds(carbohydrates fatty acids chromophores polyketides andtypes of amino acids) For each monomer chemical detailsare provided such as short name IUPAC name empiri-cal molecular formula molecular weight chemical struc-ture represented by two-dimensions chemical graphs andSMILES notation and links to PubChem (12) and theChemical Component Dictionary (14)
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1117
Efficient structure search The structure search query formhas been simplified The editor helping to draw monomericstructurecomposition of NRPs is now coded in JavaScriptThe interactivity with the user is improved with fast andsimple access to the monomer list and intuitive drawingfunctionalities A new algorithm was introduced (15) tosearch for a monomer list among the Norine peptides basedon a monomeric composition fingerprint (MCFP) Finallythe different algorithms (MCFP search pattern search andsimilarity search) were combined to output only one resultpage The peptides similar to the query are sorted by dis-tances For each found peptide a specific page shows thecomparison between it and the query This enhanced struc-ture search functionality is already operated by the anti-SMASH platform (16ndash18) When a NRPS gene cluster ispredicted a putative peptide is constructed and can be com-pared to the reference NRPs through a direct link to Norine
Filtering by graphical output The annotation or structuresearches output a result page that lists the found peptidesAn icon representing a pie chart gives access to a graphicalrepresentation Pie charts illustrate global numerical pro-portion of found peptides for several annotations The slicesare clickable and give access to the current peptide list lim-ited to the selected criteria This process allows users to filterthe results by more selective criteria
Download and customize the data Norine offers the op-portunity to download all data in several formats From apeptide or a monomer page icons give all annotations inTXTCSV HTML XML or JSON formats as illustratedin Figure 2 From the result pages listing a set of peptidesthe user can customize the annotations that will be inte-grated in the output With the HTML format a table withall the annotations selected is provided So the user cancompare the desired annotations of several peptides at atime
Norine REST web services
A web API to access the Norine database We developed aweb API (Application Programming Interface) to access theNorine database This API is based on REST (Represen-tational state transfer) REST (19) is an architectural stylebased on web-standards and the HTTP protocol Web APIsexpose resources to computer programs Whereas classicalweb approaches are mainly dedicated to humans by provid-ing rich and interactive interfaces (such as web forms or piediagrams) for querying the database Norine RESTful webservices enable integration with other resources by givingaccess to data it contains using simple resource URI Inmost of the cases data returned by such services are avail-able in various formats such as HTML XML or JSONCurrently Norine provides the following services
(i) retrieve NRP annotations from Norine ID or by name(ii) get list of all monomers with cluster tree(iii) get the access code of an external link or retrieve
database name (PDB PubChem) from access code(iv) obtain information on producing organisms by nameThe Norine REST services URIs
use the following general syntaxhttpnorinerestltpathgtltformatgt[parameter]
The path argument determines the type of service to useformat can be XML or JSON and parameter correspondsto the user query For example to retrieve annotations forthe vancomycin NRP in JSON format the URI will looklike httpnorinerestjsonnamevancomycin
Write programs for Norine REST services Another fea-ture that may interest developers of other resources anddatabases is the ability to access Norine through programsDevelopers can write client programs in any languageThese programs simply consist in creating HTTP requestswith the specific URI described above and handling the re-sponse It is also possible to embed its own HTML form toquery Norine and receive response in XML or JSON for-mats
CONCLUSION
Norine (httpbioinfoliflfrNRP) is the unique platformdedicated to NRPs A user-friendly interface eases thebrowsing annotation or structure searching and download-ing of the NRPs and their monomers This interface is com-pleted by programmatic access through a web API basedon REST Dedicated bioinformatics tools are associatedwith the database such as an editor and a visualizer of themonomeric structures homemade algorithms to comparemonomeric structures and s2m that infers monomeric struc-ture from chemical structure An important improvementis the development of MyNorine tool This tool is an in-terface to simplify the entry of new peptides or modifica-tion of existing ones in Norine by biologists The processintegrates the different steps going from the completion ofthe annotations to the validation of the submitted data Sci-entists can create an account and start entering informa-tion in MyNorine The submission will first be sent to thevalidators of Norine team before being made available inNorine Thanks to MyNorine the scientific community caneasily contribute to increase the quantity of identified NRPsand improve the quality of associated annotations storedin Norine Our goal is to initiate the crowdsourcing by ap-pealing for the authors of articles describing NRPs not inNorine to contribute to this resource The external contrib-utors will be promoted on Norine homepage and will beassociated to the NRP entries they fill in
ACKNOWLEDGEMENT
The authors would like to thank Mohcen Benmounah andAntoine Engelaere for their participation in Norine devel-opments
FUNDING
Bilille the bioinformatics service platform of Lille Univer-sity of Lille and Inria Funding for open access charge In-riaConflict of interest statement None declared
REFERENCES1 CabocheS PupinM LeclereV FontaineA JacquesP and
KucherovG (2008) NORINE a database of nonribosomal peptidesNucleic Acids Res 36 D326ndashD331
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1118 Nucleic Acids Research 2016 Vol 44 Database issue
2 NeumannNKN StoppacherN ZeilingerS DegenkolbTBrucknerH and SchuhmacherR (2015) The PeptaibioticsDatabasendashndasha comprehensive online resource Chem Biodivers 12743ndash751
3 ConwayKR and BoddyCN (2013) ClusterMine360 a database ofmicrobial PKSNRPS biosynthesis Nucleic Acids Res 41D402ndashD407
4 WalshCT (2016) Insights into the chemical logic and enzymaticmachinery of NRPS assembly lines Nat ProdRep DOI101039C5NP00035A
5 MarahielMA (2016) A structural model for multimodular NRPSassembly lines Nat Prod Rep DOI101039C5NP00082C
6 CabocheS LeclereV PupinM KucherovG and JacquesP (2010)Diversity of monomers in nonribosomal peptides towards theprediction of origin and biological activity J Bacteriol 1925143ndash5150
7 DoroghaziJR AlbrightJC GoeringAW JuK-S HainesRRTchalukovKA LabedaDP KelleherNL and MetcalfWW(2014) A roadmap for natural product discovery based on large-scalegenomics and metabolomics Nat Chem Biol 10 963ndash968
8 MedemaMH PaalvastY NguyenDD MelnikADorresteinPC TakanoE and BreitlingR (2014) Pep2Pathautomated mass spectrometry-guided genome mining of peptidicnatural products PLoS Comput Biol 10 e1003822
9 NiedermeyerTHJ and StrohalmM (2012) mMass as a softwaretool for the annotation of cyclic peptide tandem mass spectra PLoSOne 7 e44913
10 IbrahimA YangL JohnstonC LiuX MaB andMagarveyNA (2012) Dereplicating nonribosomal peptides using aninformatic search algorithm for natural products (iSNAP) discoveryProc Natl Acad Sci USA 109 19196ndash19201
11 VelankarS DanaJM JacobsenJ GinkelGv GanePJ LuoJOldfieldTJ OrsquoDonovanC MartinM-J andKleywegtGJ (2012) SIFTS structure integration with function
taxonomy and sequences resource Nucleic AcidsRes 41 D483ndashD489
12 BoltonEE WangY ThiessenPA andBryantSH (2008) PubChem Integrated Platform of SmallMolecules and Biological Activities Annu Rep ComputChem 4 217ndash241
13 BermanH HenrickK and NakamuraH (2003) Announcing theworldwide Protein Data Bank Nat Struct Mol Biol 10 980ndash980
14 WestbrookJD ShaoC FengZ ZhuravlevaM ValenkarS andYoungJ (2014) The chemical component dictionary completedescriptions of constituent molecules in experimentally determined3D macromolecules in the Protein DataBank Bioinformatics 31 1274ndash1278
15 AbdoA CabocheS LeclereV JacquesP and PupinM (2012) Anew fingerprint to predict nonribosomal peptides activity J ComputAided Mol Des 26 1187ndash1194
16 WeberT BlinK DuddelaS KrugD KimHU BruccoleriRLeeSY FischbachMA MullerR WohllebenW et al (2015)antiSMASH 30ndashndasha comprehensive resource for the genome miningof biosynthetic gene clusters Nucleic Acids Res 43 W237ndashW243
17 BlinK MedemaMH KazempourD FischbachMABreitlingR TakanoE and WeberT (2013) antiSMASH 20ndashndashaversatile platform for genome mining of secondary metaboliteproducers Nucleic Acids Res 41 W204ndashW212
19 FieldingRT and TaylorRN (2002) Principled design of themodern Web architecture ACM Trans Internet Technol 2 115ndash150
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1115
Figure 1 Overview of MyNorine the main use cases
at once Indeed users can at any time during the processdownload XML or JSON file of their current productionthen work locally on it (or not) and finally upload the file toMyNorine later In other words downloading a NRP fileenables to work off-line The submission process continuesnormally after uploading Furthermore it is possible to usean existing peptide as a template for instance to input a newvariant of a peptide already in Norine To use this featurecurators duplicate the NRP from its result page as illus-trated in Figure 2
Proposal of modification for existing entries
To improve quality of information stored in the Norinedatabase and ensure its integrity MyNorine provides a pro-cess to suggest modifications for an entry Any annotationcan be corrected or added through an icon on the peptideresult page (see Figure 2) Users can load an entry add orchange one or more data and send a notification to the val-idators with a message pointing out what has been proposedto be modified After manual check validators accept or re-ject the proposal Accepted modifications are immediatelyvisible on peptide result pages In this case the curator ismentioned as the author of the modification of the peptideentry
Validation
Manual validation Regarding validation aspectMyNorine offers to validators specific interfaces andtwo dashboards (one for new peptides the other formodifications) as illustrated in Figure 3 for new peptidesThe dashboard is divided into three tabs The first tab listscurrent unprocessed submissions and allows validators toload check accept (or reject) them The second tab listssubmissions that are under validation process The last one
shows accepted submissions and their history (submissiondate contributor)
Validators ensure that NRPs hosted in Norine are man-ually extracted from scientific knowledge (literature otherdatabases personal analysis) The NRPs that are predictedexclusively by bioinformatics tools are avoided At least oneof the following evidences is required for a submission toNorine
(i) experimental elucidation of the complete structureeventually coupled with bioinformatics analysis of the cor-responding NRPS gene cluster or
(ii) experimental mass determination coupled with thecareful bioinformatics analysis of the corresponding NRPSgene cluster or
(iii) manually curated annotation obtained by similaritywith gene clusters coding for NRP synthetases studied ex-perimentally
All the peptides submitted through MyNorine will be in-spected by validators of the Norine team Only the peptidesmatching Norine standards will be validated and incorpo-rated into the public version of the database If no evidencefor the non-ribosomal origin is presented the validatorsmay assign the putative status to the submitted peptide
Semi-automatic verifications s2m tool Manual valida-tion is strengthened by semi-automatic verifications Ba-sic checks are done automatically to ensure the consis-tency of the data within and among the entries Above allthe monomeric structures are checked through s2m tools2m infers automatically monomeric structures from chem-ical structures represented by SMILES notation Chemi-cal structures of NRPs stored in Norine are extracted fromcompound databases such as PubChem (12) or PDB (13)The predicted monomeric structures are compared to theNorine structures We detected 97 incongruities between thestructures in Norine and the ones of other databases A re-
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1116 Nucleic Acids Research 2016 Vol 44 Database issue
Figure 2 Duplicate an existing NRP or propose modification
Figure 3 Dashboard for validators
mediation process started to identify from which databasescomes each error and to correct the corresponding annota-tions
FACILITATION OF NORINE DATA ACCESS
Since the creation of Norine the interface has been im-proved to ease the usage of the website the search for rele-vant data and their visualization The major improvementssince the 2008 publication are presented in the followingparagraphs
Improved Norine user interface
Access to monomers annotations A query form enablesto search for the 529 monomers constituting the NRPsstored in Norine They are clustered by class of compounds(carbohydrates fatty acids chromophores polyketides andtypes of amino acids) For each monomer chemical detailsare provided such as short name IUPAC name empiri-cal molecular formula molecular weight chemical struc-ture represented by two-dimensions chemical graphs andSMILES notation and links to PubChem (12) and theChemical Component Dictionary (14)
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1117
Efficient structure search The structure search query formhas been simplified The editor helping to draw monomericstructurecomposition of NRPs is now coded in JavaScriptThe interactivity with the user is improved with fast andsimple access to the monomer list and intuitive drawingfunctionalities A new algorithm was introduced (15) tosearch for a monomer list among the Norine peptides basedon a monomeric composition fingerprint (MCFP) Finallythe different algorithms (MCFP search pattern search andsimilarity search) were combined to output only one resultpage The peptides similar to the query are sorted by dis-tances For each found peptide a specific page shows thecomparison between it and the query This enhanced struc-ture search functionality is already operated by the anti-SMASH platform (16ndash18) When a NRPS gene cluster ispredicted a putative peptide is constructed and can be com-pared to the reference NRPs through a direct link to Norine
Filtering by graphical output The annotation or structuresearches output a result page that lists the found peptidesAn icon representing a pie chart gives access to a graphicalrepresentation Pie charts illustrate global numerical pro-portion of found peptides for several annotations The slicesare clickable and give access to the current peptide list lim-ited to the selected criteria This process allows users to filterthe results by more selective criteria
Download and customize the data Norine offers the op-portunity to download all data in several formats From apeptide or a monomer page icons give all annotations inTXTCSV HTML XML or JSON formats as illustratedin Figure 2 From the result pages listing a set of peptidesthe user can customize the annotations that will be inte-grated in the output With the HTML format a table withall the annotations selected is provided So the user cancompare the desired annotations of several peptides at atime
Norine REST web services
A web API to access the Norine database We developed aweb API (Application Programming Interface) to access theNorine database This API is based on REST (Represen-tational state transfer) REST (19) is an architectural stylebased on web-standards and the HTTP protocol Web APIsexpose resources to computer programs Whereas classicalweb approaches are mainly dedicated to humans by provid-ing rich and interactive interfaces (such as web forms or piediagrams) for querying the database Norine RESTful webservices enable integration with other resources by givingaccess to data it contains using simple resource URI Inmost of the cases data returned by such services are avail-able in various formats such as HTML XML or JSONCurrently Norine provides the following services
(i) retrieve NRP annotations from Norine ID or by name(ii) get list of all monomers with cluster tree(iii) get the access code of an external link or retrieve
database name (PDB PubChem) from access code(iv) obtain information on producing organisms by nameThe Norine REST services URIs
use the following general syntaxhttpnorinerestltpathgtltformatgt[parameter]
The path argument determines the type of service to useformat can be XML or JSON and parameter correspondsto the user query For example to retrieve annotations forthe vancomycin NRP in JSON format the URI will looklike httpnorinerestjsonnamevancomycin
Write programs for Norine REST services Another fea-ture that may interest developers of other resources anddatabases is the ability to access Norine through programsDevelopers can write client programs in any languageThese programs simply consist in creating HTTP requestswith the specific URI described above and handling the re-sponse It is also possible to embed its own HTML form toquery Norine and receive response in XML or JSON for-mats
CONCLUSION
Norine (httpbioinfoliflfrNRP) is the unique platformdedicated to NRPs A user-friendly interface eases thebrowsing annotation or structure searching and download-ing of the NRPs and their monomers This interface is com-pleted by programmatic access through a web API basedon REST Dedicated bioinformatics tools are associatedwith the database such as an editor and a visualizer of themonomeric structures homemade algorithms to comparemonomeric structures and s2m that infers monomeric struc-ture from chemical structure An important improvementis the development of MyNorine tool This tool is an in-terface to simplify the entry of new peptides or modifica-tion of existing ones in Norine by biologists The processintegrates the different steps going from the completion ofthe annotations to the validation of the submitted data Sci-entists can create an account and start entering informa-tion in MyNorine The submission will first be sent to thevalidators of Norine team before being made available inNorine Thanks to MyNorine the scientific community caneasily contribute to increase the quantity of identified NRPsand improve the quality of associated annotations storedin Norine Our goal is to initiate the crowdsourcing by ap-pealing for the authors of articles describing NRPs not inNorine to contribute to this resource The external contrib-utors will be promoted on Norine homepage and will beassociated to the NRP entries they fill in
ACKNOWLEDGEMENT
The authors would like to thank Mohcen Benmounah andAntoine Engelaere for their participation in Norine devel-opments
FUNDING
Bilille the bioinformatics service platform of Lille Univer-sity of Lille and Inria Funding for open access charge In-riaConflict of interest statement None declared
REFERENCES1 CabocheS PupinM LeclereV FontaineA JacquesP and
KucherovG (2008) NORINE a database of nonribosomal peptidesNucleic Acids Res 36 D326ndashD331
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1118 Nucleic Acids Research 2016 Vol 44 Database issue
2 NeumannNKN StoppacherN ZeilingerS DegenkolbTBrucknerH and SchuhmacherR (2015) The PeptaibioticsDatabasendashndasha comprehensive online resource Chem Biodivers 12743ndash751
3 ConwayKR and BoddyCN (2013) ClusterMine360 a database ofmicrobial PKSNRPS biosynthesis Nucleic Acids Res 41D402ndashD407
4 WalshCT (2016) Insights into the chemical logic and enzymaticmachinery of NRPS assembly lines Nat ProdRep DOI101039C5NP00035A
5 MarahielMA (2016) A structural model for multimodular NRPSassembly lines Nat Prod Rep DOI101039C5NP00082C
6 CabocheS LeclereV PupinM KucherovG and JacquesP (2010)Diversity of monomers in nonribosomal peptides towards theprediction of origin and biological activity J Bacteriol 1925143ndash5150
7 DoroghaziJR AlbrightJC GoeringAW JuK-S HainesRRTchalukovKA LabedaDP KelleherNL and MetcalfWW(2014) A roadmap for natural product discovery based on large-scalegenomics and metabolomics Nat Chem Biol 10 963ndash968
8 MedemaMH PaalvastY NguyenDD MelnikADorresteinPC TakanoE and BreitlingR (2014) Pep2Pathautomated mass spectrometry-guided genome mining of peptidicnatural products PLoS Comput Biol 10 e1003822
9 NiedermeyerTHJ and StrohalmM (2012) mMass as a softwaretool for the annotation of cyclic peptide tandem mass spectra PLoSOne 7 e44913
10 IbrahimA YangL JohnstonC LiuX MaB andMagarveyNA (2012) Dereplicating nonribosomal peptides using aninformatic search algorithm for natural products (iSNAP) discoveryProc Natl Acad Sci USA 109 19196ndash19201
11 VelankarS DanaJM JacobsenJ GinkelGv GanePJ LuoJOldfieldTJ OrsquoDonovanC MartinM-J andKleywegtGJ (2012) SIFTS structure integration with function
taxonomy and sequences resource Nucleic AcidsRes 41 D483ndashD489
12 BoltonEE WangY ThiessenPA andBryantSH (2008) PubChem Integrated Platform of SmallMolecules and Biological Activities Annu Rep ComputChem 4 217ndash241
13 BermanH HenrickK and NakamuraH (2003) Announcing theworldwide Protein Data Bank Nat Struct Mol Biol 10 980ndash980
14 WestbrookJD ShaoC FengZ ZhuravlevaM ValenkarS andYoungJ (2014) The chemical component dictionary completedescriptions of constituent molecules in experimentally determined3D macromolecules in the Protein DataBank Bioinformatics 31 1274ndash1278
15 AbdoA CabocheS LeclereV JacquesP and PupinM (2012) Anew fingerprint to predict nonribosomal peptides activity J ComputAided Mol Des 26 1187ndash1194
16 WeberT BlinK DuddelaS KrugD KimHU BruccoleriRLeeSY FischbachMA MullerR WohllebenW et al (2015)antiSMASH 30ndashndasha comprehensive resource for the genome miningof biosynthetic gene clusters Nucleic Acids Res 43 W237ndashW243
17 BlinK MedemaMH KazempourD FischbachMABreitlingR TakanoE and WeberT (2013) antiSMASH 20ndashndashaversatile platform for genome mining of secondary metaboliteproducers Nucleic Acids Res 41 W204ndashW212
19 FieldingRT and TaylorRN (2002) Principled design of themodern Web architecture ACM Trans Internet Technol 2 115ndash150
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1116 Nucleic Acids Research 2016 Vol 44 Database issue
Figure 2 Duplicate an existing NRP or propose modification
Figure 3 Dashboard for validators
mediation process started to identify from which databasescomes each error and to correct the corresponding annota-tions
FACILITATION OF NORINE DATA ACCESS
Since the creation of Norine the interface has been im-proved to ease the usage of the website the search for rele-vant data and their visualization The major improvementssince the 2008 publication are presented in the followingparagraphs
Improved Norine user interface
Access to monomers annotations A query form enablesto search for the 529 monomers constituting the NRPsstored in Norine They are clustered by class of compounds(carbohydrates fatty acids chromophores polyketides andtypes of amino acids) For each monomer chemical detailsare provided such as short name IUPAC name empiri-cal molecular formula molecular weight chemical struc-ture represented by two-dimensions chemical graphs andSMILES notation and links to PubChem (12) and theChemical Component Dictionary (14)
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1117
Efficient structure search The structure search query formhas been simplified The editor helping to draw monomericstructurecomposition of NRPs is now coded in JavaScriptThe interactivity with the user is improved with fast andsimple access to the monomer list and intuitive drawingfunctionalities A new algorithm was introduced (15) tosearch for a monomer list among the Norine peptides basedon a monomeric composition fingerprint (MCFP) Finallythe different algorithms (MCFP search pattern search andsimilarity search) were combined to output only one resultpage The peptides similar to the query are sorted by dis-tances For each found peptide a specific page shows thecomparison between it and the query This enhanced struc-ture search functionality is already operated by the anti-SMASH platform (16ndash18) When a NRPS gene cluster ispredicted a putative peptide is constructed and can be com-pared to the reference NRPs through a direct link to Norine
Filtering by graphical output The annotation or structuresearches output a result page that lists the found peptidesAn icon representing a pie chart gives access to a graphicalrepresentation Pie charts illustrate global numerical pro-portion of found peptides for several annotations The slicesare clickable and give access to the current peptide list lim-ited to the selected criteria This process allows users to filterthe results by more selective criteria
Download and customize the data Norine offers the op-portunity to download all data in several formats From apeptide or a monomer page icons give all annotations inTXTCSV HTML XML or JSON formats as illustratedin Figure 2 From the result pages listing a set of peptidesthe user can customize the annotations that will be inte-grated in the output With the HTML format a table withall the annotations selected is provided So the user cancompare the desired annotations of several peptides at atime
Norine REST web services
A web API to access the Norine database We developed aweb API (Application Programming Interface) to access theNorine database This API is based on REST (Represen-tational state transfer) REST (19) is an architectural stylebased on web-standards and the HTTP protocol Web APIsexpose resources to computer programs Whereas classicalweb approaches are mainly dedicated to humans by provid-ing rich and interactive interfaces (such as web forms or piediagrams) for querying the database Norine RESTful webservices enable integration with other resources by givingaccess to data it contains using simple resource URI Inmost of the cases data returned by such services are avail-able in various formats such as HTML XML or JSONCurrently Norine provides the following services
(i) retrieve NRP annotations from Norine ID or by name(ii) get list of all monomers with cluster tree(iii) get the access code of an external link or retrieve
database name (PDB PubChem) from access code(iv) obtain information on producing organisms by nameThe Norine REST services URIs
use the following general syntaxhttpnorinerestltpathgtltformatgt[parameter]
The path argument determines the type of service to useformat can be XML or JSON and parameter correspondsto the user query For example to retrieve annotations forthe vancomycin NRP in JSON format the URI will looklike httpnorinerestjsonnamevancomycin
Write programs for Norine REST services Another fea-ture that may interest developers of other resources anddatabases is the ability to access Norine through programsDevelopers can write client programs in any languageThese programs simply consist in creating HTTP requestswith the specific URI described above and handling the re-sponse It is also possible to embed its own HTML form toquery Norine and receive response in XML or JSON for-mats
CONCLUSION
Norine (httpbioinfoliflfrNRP) is the unique platformdedicated to NRPs A user-friendly interface eases thebrowsing annotation or structure searching and download-ing of the NRPs and their monomers This interface is com-pleted by programmatic access through a web API basedon REST Dedicated bioinformatics tools are associatedwith the database such as an editor and a visualizer of themonomeric structures homemade algorithms to comparemonomeric structures and s2m that infers monomeric struc-ture from chemical structure An important improvementis the development of MyNorine tool This tool is an in-terface to simplify the entry of new peptides or modifica-tion of existing ones in Norine by biologists The processintegrates the different steps going from the completion ofthe annotations to the validation of the submitted data Sci-entists can create an account and start entering informa-tion in MyNorine The submission will first be sent to thevalidators of Norine team before being made available inNorine Thanks to MyNorine the scientific community caneasily contribute to increase the quantity of identified NRPsand improve the quality of associated annotations storedin Norine Our goal is to initiate the crowdsourcing by ap-pealing for the authors of articles describing NRPs not inNorine to contribute to this resource The external contrib-utors will be promoted on Norine homepage and will beassociated to the NRP entries they fill in
ACKNOWLEDGEMENT
The authors would like to thank Mohcen Benmounah andAntoine Engelaere for their participation in Norine devel-opments
FUNDING
Bilille the bioinformatics service platform of Lille Univer-sity of Lille and Inria Funding for open access charge In-riaConflict of interest statement None declared
REFERENCES1 CabocheS PupinM LeclereV FontaineA JacquesP and
KucherovG (2008) NORINE a database of nonribosomal peptidesNucleic Acids Res 36 D326ndashD331
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1118 Nucleic Acids Research 2016 Vol 44 Database issue
2 NeumannNKN StoppacherN ZeilingerS DegenkolbTBrucknerH and SchuhmacherR (2015) The PeptaibioticsDatabasendashndasha comprehensive online resource Chem Biodivers 12743ndash751
3 ConwayKR and BoddyCN (2013) ClusterMine360 a database ofmicrobial PKSNRPS biosynthesis Nucleic Acids Res 41D402ndashD407
4 WalshCT (2016) Insights into the chemical logic and enzymaticmachinery of NRPS assembly lines Nat ProdRep DOI101039C5NP00035A
5 MarahielMA (2016) A structural model for multimodular NRPSassembly lines Nat Prod Rep DOI101039C5NP00082C
6 CabocheS LeclereV PupinM KucherovG and JacquesP (2010)Diversity of monomers in nonribosomal peptides towards theprediction of origin and biological activity J Bacteriol 1925143ndash5150
7 DoroghaziJR AlbrightJC GoeringAW JuK-S HainesRRTchalukovKA LabedaDP KelleherNL and MetcalfWW(2014) A roadmap for natural product discovery based on large-scalegenomics and metabolomics Nat Chem Biol 10 963ndash968
8 MedemaMH PaalvastY NguyenDD MelnikADorresteinPC TakanoE and BreitlingR (2014) Pep2Pathautomated mass spectrometry-guided genome mining of peptidicnatural products PLoS Comput Biol 10 e1003822
9 NiedermeyerTHJ and StrohalmM (2012) mMass as a softwaretool for the annotation of cyclic peptide tandem mass spectra PLoSOne 7 e44913
10 IbrahimA YangL JohnstonC LiuX MaB andMagarveyNA (2012) Dereplicating nonribosomal peptides using aninformatic search algorithm for natural products (iSNAP) discoveryProc Natl Acad Sci USA 109 19196ndash19201
11 VelankarS DanaJM JacobsenJ GinkelGv GanePJ LuoJOldfieldTJ OrsquoDonovanC MartinM-J andKleywegtGJ (2012) SIFTS structure integration with function
taxonomy and sequences resource Nucleic AcidsRes 41 D483ndashD489
12 BoltonEE WangY ThiessenPA andBryantSH (2008) PubChem Integrated Platform of SmallMolecules and Biological Activities Annu Rep ComputChem 4 217ndash241
13 BermanH HenrickK and NakamuraH (2003) Announcing theworldwide Protein Data Bank Nat Struct Mol Biol 10 980ndash980
14 WestbrookJD ShaoC FengZ ZhuravlevaM ValenkarS andYoungJ (2014) The chemical component dictionary completedescriptions of constituent molecules in experimentally determined3D macromolecules in the Protein DataBank Bioinformatics 31 1274ndash1278
15 AbdoA CabocheS LeclereV JacquesP and PupinM (2012) Anew fingerprint to predict nonribosomal peptides activity J ComputAided Mol Des 26 1187ndash1194
16 WeberT BlinK DuddelaS KrugD KimHU BruccoleriRLeeSY FischbachMA MullerR WohllebenW et al (2015)antiSMASH 30ndashndasha comprehensive resource for the genome miningof biosynthetic gene clusters Nucleic Acids Res 43 W237ndashW243
17 BlinK MedemaMH KazempourD FischbachMABreitlingR TakanoE and WeberT (2013) antiSMASH 20ndashndashaversatile platform for genome mining of secondary metaboliteproducers Nucleic Acids Res 41 W204ndashW212
19 FieldingRT and TaylorRN (2002) Principled design of themodern Web architecture ACM Trans Internet Technol 2 115ndash150
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
Nucleic Acids Research 2016 Vol 44 Database issue D1117
Efficient structure search The structure search query formhas been simplified The editor helping to draw monomericstructurecomposition of NRPs is now coded in JavaScriptThe interactivity with the user is improved with fast andsimple access to the monomer list and intuitive drawingfunctionalities A new algorithm was introduced (15) tosearch for a monomer list among the Norine peptides basedon a monomeric composition fingerprint (MCFP) Finallythe different algorithms (MCFP search pattern search andsimilarity search) were combined to output only one resultpage The peptides similar to the query are sorted by dis-tances For each found peptide a specific page shows thecomparison between it and the query This enhanced struc-ture search functionality is already operated by the anti-SMASH platform (16ndash18) When a NRPS gene cluster ispredicted a putative peptide is constructed and can be com-pared to the reference NRPs through a direct link to Norine
Filtering by graphical output The annotation or structuresearches output a result page that lists the found peptidesAn icon representing a pie chart gives access to a graphicalrepresentation Pie charts illustrate global numerical pro-portion of found peptides for several annotations The slicesare clickable and give access to the current peptide list lim-ited to the selected criteria This process allows users to filterthe results by more selective criteria
Download and customize the data Norine offers the op-portunity to download all data in several formats From apeptide or a monomer page icons give all annotations inTXTCSV HTML XML or JSON formats as illustratedin Figure 2 From the result pages listing a set of peptidesthe user can customize the annotations that will be inte-grated in the output With the HTML format a table withall the annotations selected is provided So the user cancompare the desired annotations of several peptides at atime
Norine REST web services
A web API to access the Norine database We developed aweb API (Application Programming Interface) to access theNorine database This API is based on REST (Represen-tational state transfer) REST (19) is an architectural stylebased on web-standards and the HTTP protocol Web APIsexpose resources to computer programs Whereas classicalweb approaches are mainly dedicated to humans by provid-ing rich and interactive interfaces (such as web forms or piediagrams) for querying the database Norine RESTful webservices enable integration with other resources by givingaccess to data it contains using simple resource URI Inmost of the cases data returned by such services are avail-able in various formats such as HTML XML or JSONCurrently Norine provides the following services
(i) retrieve NRP annotations from Norine ID or by name(ii) get list of all monomers with cluster tree(iii) get the access code of an external link or retrieve
database name (PDB PubChem) from access code(iv) obtain information on producing organisms by nameThe Norine REST services URIs
use the following general syntaxhttpnorinerestltpathgtltformatgt[parameter]
The path argument determines the type of service to useformat can be XML or JSON and parameter correspondsto the user query For example to retrieve annotations forthe vancomycin NRP in JSON format the URI will looklike httpnorinerestjsonnamevancomycin
Write programs for Norine REST services Another fea-ture that may interest developers of other resources anddatabases is the ability to access Norine through programsDevelopers can write client programs in any languageThese programs simply consist in creating HTTP requestswith the specific URI described above and handling the re-sponse It is also possible to embed its own HTML form toquery Norine and receive response in XML or JSON for-mats
CONCLUSION
Norine (httpbioinfoliflfrNRP) is the unique platformdedicated to NRPs A user-friendly interface eases thebrowsing annotation or structure searching and download-ing of the NRPs and their monomers This interface is com-pleted by programmatic access through a web API basedon REST Dedicated bioinformatics tools are associatedwith the database such as an editor and a visualizer of themonomeric structures homemade algorithms to comparemonomeric structures and s2m that infers monomeric struc-ture from chemical structure An important improvementis the development of MyNorine tool This tool is an in-terface to simplify the entry of new peptides or modifica-tion of existing ones in Norine by biologists The processintegrates the different steps going from the completion ofthe annotations to the validation of the submitted data Sci-entists can create an account and start entering informa-tion in MyNorine The submission will first be sent to thevalidators of Norine team before being made available inNorine Thanks to MyNorine the scientific community caneasily contribute to increase the quantity of identified NRPsand improve the quality of associated annotations storedin Norine Our goal is to initiate the crowdsourcing by ap-pealing for the authors of articles describing NRPs not inNorine to contribute to this resource The external contrib-utors will be promoted on Norine homepage and will beassociated to the NRP entries they fill in
ACKNOWLEDGEMENT
The authors would like to thank Mohcen Benmounah andAntoine Engelaere for their participation in Norine devel-opments
FUNDING
Bilille the bioinformatics service platform of Lille Univer-sity of Lille and Inria Funding for open access charge In-riaConflict of interest statement None declared
REFERENCES1 CabocheS PupinM LeclereV FontaineA JacquesP and
KucherovG (2008) NORINE a database of nonribosomal peptidesNucleic Acids Res 36 D326ndashD331
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1118 Nucleic Acids Research 2016 Vol 44 Database issue
2 NeumannNKN StoppacherN ZeilingerS DegenkolbTBrucknerH and SchuhmacherR (2015) The PeptaibioticsDatabasendashndasha comprehensive online resource Chem Biodivers 12743ndash751
3 ConwayKR and BoddyCN (2013) ClusterMine360 a database ofmicrobial PKSNRPS biosynthesis Nucleic Acids Res 41D402ndashD407
4 WalshCT (2016) Insights into the chemical logic and enzymaticmachinery of NRPS assembly lines Nat ProdRep DOI101039C5NP00035A
5 MarahielMA (2016) A structural model for multimodular NRPSassembly lines Nat Prod Rep DOI101039C5NP00082C
6 CabocheS LeclereV PupinM KucherovG and JacquesP (2010)Diversity of monomers in nonribosomal peptides towards theprediction of origin and biological activity J Bacteriol 1925143ndash5150
7 DoroghaziJR AlbrightJC GoeringAW JuK-S HainesRRTchalukovKA LabedaDP KelleherNL and MetcalfWW(2014) A roadmap for natural product discovery based on large-scalegenomics and metabolomics Nat Chem Biol 10 963ndash968
8 MedemaMH PaalvastY NguyenDD MelnikADorresteinPC TakanoE and BreitlingR (2014) Pep2Pathautomated mass spectrometry-guided genome mining of peptidicnatural products PLoS Comput Biol 10 e1003822
9 NiedermeyerTHJ and StrohalmM (2012) mMass as a softwaretool for the annotation of cyclic peptide tandem mass spectra PLoSOne 7 e44913
10 IbrahimA YangL JohnstonC LiuX MaB andMagarveyNA (2012) Dereplicating nonribosomal peptides using aninformatic search algorithm for natural products (iSNAP) discoveryProc Natl Acad Sci USA 109 19196ndash19201
11 VelankarS DanaJM JacobsenJ GinkelGv GanePJ LuoJOldfieldTJ OrsquoDonovanC MartinM-J andKleywegtGJ (2012) SIFTS structure integration with function
taxonomy and sequences resource Nucleic AcidsRes 41 D483ndashD489
12 BoltonEE WangY ThiessenPA andBryantSH (2008) PubChem Integrated Platform of SmallMolecules and Biological Activities Annu Rep ComputChem 4 217ndash241
13 BermanH HenrickK and NakamuraH (2003) Announcing theworldwide Protein Data Bank Nat Struct Mol Biol 10 980ndash980
14 WestbrookJD ShaoC FengZ ZhuravlevaM ValenkarS andYoungJ (2014) The chemical component dictionary completedescriptions of constituent molecules in experimentally determined3D macromolecules in the Protein DataBank Bioinformatics 31 1274ndash1278
15 AbdoA CabocheS LeclereV JacquesP and PupinM (2012) Anew fingerprint to predict nonribosomal peptides activity J ComputAided Mol Des 26 1187ndash1194
16 WeberT BlinK DuddelaS KrugD KimHU BruccoleriRLeeSY FischbachMA MullerR WohllebenW et al (2015)antiSMASH 30ndashndasha comprehensive resource for the genome miningof biosynthetic gene clusters Nucleic Acids Res 43 W237ndashW243
17 BlinK MedemaMH KazempourD FischbachMABreitlingR TakanoE and WeberT (2013) antiSMASH 20ndashndashaversatile platform for genome mining of secondary metaboliteproducers Nucleic Acids Res 41 W204ndashW212
19 FieldingRT and TaylorRN (2002) Principled design of themodern Web architecture ACM Trans Internet Technol 2 115ndash150
Downloaded from httpsacademicoupcomnararticle-abstract44D1D11132502639by gueston 15 February 2018
D1118 Nucleic Acids Research 2016 Vol 44 Database issue
2 NeumannNKN StoppacherN ZeilingerS DegenkolbTBrucknerH and SchuhmacherR (2015) The PeptaibioticsDatabasendashndasha comprehensive online resource Chem Biodivers 12743ndash751
3 ConwayKR and BoddyCN (2013) ClusterMine360 a database ofmicrobial PKSNRPS biosynthesis Nucleic Acids Res 41D402ndashD407
4 WalshCT (2016) Insights into the chemical logic and enzymaticmachinery of NRPS assembly lines Nat ProdRep DOI101039C5NP00035A
5 MarahielMA (2016) A structural model for multimodular NRPSassembly lines Nat Prod Rep DOI101039C5NP00082C
6 CabocheS LeclereV PupinM KucherovG and JacquesP (2010)Diversity of monomers in nonribosomal peptides towards theprediction of origin and biological activity J Bacteriol 1925143ndash5150
7 DoroghaziJR AlbrightJC GoeringAW JuK-S HainesRRTchalukovKA LabedaDP KelleherNL and MetcalfWW(2014) A roadmap for natural product discovery based on large-scalegenomics and metabolomics Nat Chem Biol 10 963ndash968
8 MedemaMH PaalvastY NguyenDD MelnikADorresteinPC TakanoE and BreitlingR (2014) Pep2Pathautomated mass spectrometry-guided genome mining of peptidicnatural products PLoS Comput Biol 10 e1003822
9 NiedermeyerTHJ and StrohalmM (2012) mMass as a softwaretool for the annotation of cyclic peptide tandem mass spectra PLoSOne 7 e44913
10 IbrahimA YangL JohnstonC LiuX MaB andMagarveyNA (2012) Dereplicating nonribosomal peptides using aninformatic search algorithm for natural products (iSNAP) discoveryProc Natl Acad Sci USA 109 19196ndash19201
11 VelankarS DanaJM JacobsenJ GinkelGv GanePJ LuoJOldfieldTJ OrsquoDonovanC MartinM-J andKleywegtGJ (2012) SIFTS structure integration with function
taxonomy and sequences resource Nucleic AcidsRes 41 D483ndashD489
12 BoltonEE WangY ThiessenPA andBryantSH (2008) PubChem Integrated Platform of SmallMolecules and Biological Activities Annu Rep ComputChem 4 217ndash241
13 BermanH HenrickK and NakamuraH (2003) Announcing theworldwide Protein Data Bank Nat Struct Mol Biol 10 980ndash980
14 WestbrookJD ShaoC FengZ ZhuravlevaM ValenkarS andYoungJ (2014) The chemical component dictionary completedescriptions of constituent molecules in experimentally determined3D macromolecules in the Protein DataBank Bioinformatics 31 1274ndash1278
15 AbdoA CabocheS LeclereV JacquesP and PupinM (2012) Anew fingerprint to predict nonribosomal peptides activity J ComputAided Mol Des 26 1187ndash1194
16 WeberT BlinK DuddelaS KrugD KimHU BruccoleriRLeeSY FischbachMA MullerR WohllebenW et al (2015)antiSMASH 30ndashndasha comprehensive resource for the genome miningof biosynthetic gene clusters Nucleic Acids Res 43 W237ndashW243
17 BlinK MedemaMH KazempourD FischbachMABreitlingR TakanoE and WeberT (2013) antiSMASH 20ndashndashaversatile platform for genome mining of secondary metaboliteproducers Nucleic Acids Res 41 W204ndashW212
![PHD KnowledgeBase Software User's Guide KnowledgeBase...knowledgebase file, testing the connection to the PHD-ODE server from which resources are downloaded [PHD-ODE only], displaying](https://static.documents.pub/doc/80x56/5f62ef8c765e9d699e179689/phd-knowledgebase-software-users-knowledgebase-knowledgebase-file-testing-the.jpg)
