the patentability of biomolecules – does online...

Review

The patentability of biomolecules – doesonline bioinformatics compromise novelty?

Anton Hutter*Appleyard Lees, Chartered and European Patent Attorneys, 15 Clare Road, Halifax, HX1 2HY, UK

*Correspondence to:Appleyard Lees, Chartered andEuropean Patent Attorneys, 15Clare Road, Halifax, HX1 2HY,UK.E-mail:[email protected]

Received: 23 January 2002

Accepted: 6 February 2002

Abstract

Researchers are becoming increasingly concerned that the confidentiality of their novel

biomolecule sequences is being jeopardised, particularly when these sequences are either

submitted to sequence databases or uploaded as query terms onto internet-based

bioinformatic software suites. The researcher’s fears stem from the fact that the actual

uploading of their sequences acts as a novelty destroying prior disclosure or publication,

and that this may subsequently preclude valid patent protection for the sequences. This

article addresses the key issues involved in the analyses of biomolecules, highlighting

potential risks taken by many researchers in regard to patent protection and suggests

possible ways in which these risks may be mitigated. Copyright # 2002 John Wiley &

Sons, Ltd.

Keywords: Biomolecule; Bioinformatics; gene sequence; patent; confidential; novelty;

disclosure; Internet; e-mail

The requirements of patentability inEurope

In order to obtain a valid European patent for anyinvention, the invention must be novel (Art.54European Patent Convention; EPC), exhibit aninventive step (Art.56 EPC), and be susceptible ofindustrial application (Art.57 EPC). An invention isregarded as being novel if it does not form part ofthe state of the art. The state of the art consists ofeverything made available to the public anywhere inthe world prior to the filing date of a Europeanpatent application. An invention is taken to involvean inventive step if, when judged by a person skilledin the art at the time of filing the patent application,it is not merely an obvious development of the stateof the art. Finally, an invention is considered asbeing susceptible of industrial application if it canbe made or used in any kind of industry, includingagriculture.

In order to obtain a patent directed to a newlydiscovered biomolecule, for example, a sequence orpartial sequence of a gene or protein, one facet ofnovelty is that the biomolecule must be in anisolated form (Art.5(2) European BiotechnologyDirective; BD). In addition, in order to meet the

industrial applicability requirement, a specific utilityfor the sequence or partial sequence which isbeyond the realm of mere speculation must bedisclosed at the filing date of the patent application(Art.5(3) BD). Hence, it is determining whichspecific parts of a genome constitute functionalgenes, and what the functions of those genes are,that may lead to patentable inventions.

At a time when the genome projects of numerousorganisms are nearing completion and, in manycases, where the genomes have been fullysequenced, a large amount of biomolecular data isbeing constantly generated at an incredible rate.Consequently, researchers are filing patent applica-tions for all types of biomolecules in an attempt toprotect their intellectual property stemming fromthis mass of genomic data. In order for these inven-tions to form the subject of valid patents, they mustmeet all the requirements of patentability men-tioned above. Therefore, when assessing whether abiomolecule is patentable, one of the first questionsthat must be asked is: ‘Is it new?’ The question ofnovelty is very strict and, as part of the requirementfor novelty, the biomolecule must be confidentialwhen a patent application in respect of that bio-molecule is filed. If the sequence of the biomolecule

Comparative and Functional Genomics

Comp Funct Genom 2002; 3: 119–126.Published online 1 March 2002 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002 /cfg.149

Copyright # 2002 John Wiley & Sons, Ltd.

has been publicly used or disclosed, or made availa-ble to the public in any way whatsoever prior to thefiling date of the patent application, then its noveltymay be in doubt. Hence, the validity of the pro-spective patent may be at risk.

As part of the process of genome research,specifically genome sequencing, the nucleic acidsequences of newly discovered genes are normallymade publicly available as cDNAs, via online data-bases such as those of the European MolecularBiology Laboratory (EMBL), and GenBank. Inaddition, newly discovered proteins are often pub-lished via online databases such as Swiss-PROT andthe Martinsried Institute for Protein Sequences(MIPS). GenBank, the EMBL Nucleotide SequenceDatabase and the DNA Databank of Japan(DDBJ) provide those submitting a sequence totheir database with the option of specifying that thesequence data is not added to the online database,and is held confidential for a specified period oftime (a ‘holding period’). For example, the sub-mitter may require that the sequence is added to theonline database only after the sequence has beenpublished in a journal, or after a patent applicationhas been filed in respect of that sequence, therebymaintaining its confidentiality. Around 70–80% ofnew sequences submitted to the EMBL nucleotidedatabase are subjected to such a confidential hold-ing period and neither the EBI nor the EPO regardthis as a novelty destroying public disclosure (PeterStoehr, personal communication).

Prior to filing a patent application directed to abiomolecule such as a nucleic acid or protein, it isoften advisable to investigate the novelty of thesequence of that biomolecule to avoid filing anapplication for a biomolecule that has already beenpublicly disclosed, possibly via one of these onlinedatabases, or otherwise. The European Patent Orga-nisation (EPO) provides various search facilities forbiomolecules by which these investigations may becarried out (http://www.european-patent-office.org/dg1/ssp/html). The EPO website includes a clearstatement that sequence searching carried out bythe EPO is strictly confidential and therefore doesnot risk the novelty of the sequences searched interms of patentability.

The purpose of this paper is to investigate thenovelty of biomolecules forming the subject matterof patents and patent applications. Specifically, itanalyses the use of online bioinformatics softwareby researchers to determine putative structure andfunction of their newly sequenced biomolecules, and

whether such use online amounts to a prior publicdisclosure of that biomolecule. As noted above,such a public disclosure may well invalidate a laterfiled patent application directed to that biomole-cule.

Biology using The Internet

Online bioinformatics is carried out via a networkof computers as illustrated by Boxes A-D inFigure 1. The sequence of a biomolecule which, forthe purposes of this paper, is a newly sequencednucleic acid denoted by the term SEQ ID, is storedon a researcher’s computer, or client (Box A). Forexample, SEQ ID may be a gene encoding a novelpolypeptide and is regarded as being confidentialsince it is within the confines of the researcher’scomputer. In addition, it is assumed that thesequence has not been published in an academicpaper, publicly used or disclosed to third parties bythe researcher in any way. Hence, while thesequence is still on the researcher’s computer, it isstill novel in terms of the patenting requirements. Inan attempt to determine whether SEQ ID isexpressed, the researcher’s next step might be toupload the sequence as a query term for subsequentbioinformatics analyses using the wide range ofstructural and functional predictive software whichare currently freely available online via a softwareserver (Box C), for example, the National Centrefor Biotechnology Information (NCBI), the HumanGenome Mapping Project – Resource Centre(HGMP-RC), or the European BioinformaticsInstitute (EBI) etc.

In the case of large pharmaceutical companiesand universities, the researcher’s computer is able tocontact the bioinformatics software server directlyvia an in-house Internet gateway. However, in thecase of smaller organisations, or in circumstanceswhere the researcher accesses the software serveronline from home, the researcher’s computer con-tacts the software server via a commercial InternetService Provider (ISP) (Box B), for example,Demon or AOL etc. In the latter case, the queryterm is transferred to the ISP as a single packet ofdata via a telephone line shown as path (i). Thequery term is then divided up and sent as a numberof packets of data by the commercial ISP alongpath (ii) to the software server where the packetsof data are reassembled to form the completequery term, SEQ ID. The server, which provides

120 A. Hutter

Copyright # 2002 John Wiley & Sons, Ltd. Comp Funct Genom 2002; 3: 119–126.

bioinformatics software such as the Basic LocalAlignment Search Tool (BLAST) program, thendivides the query term up into a new set of datapackets and sends these along path (iii) to anotherserver (Box D). This second server, for example,EMBL, provides an external database of sequencesagainst which the query term, SEQ ID, may besearched.

Hence, by carrying out bioinformatics analysis ofSEQ ID online via the internet, the researcher hasvoluntarily uploaded the confidential query termfrom his computer (Box A) onto a bioinformaticssoftware server (Box C), and, therefore, permittedthe software server to search the sequence against asequence database (Box D). The format of resultsgenerated by bioinformatics analysis varies widelydepending on the predictive software and serverused, but in the case of the BLAST program, inmost cases the results consist of a detailed list ofhomology scores and alignments with other seq-uences in the database, which together may be usedto infer putative structure and function of the querysequence, SEQ ID. Such data generated in silico isoften included in patent applications in an attemptto meet the utility requirement mentioned above.The results are delivered back to the researcherfrom Box D to Box A along paths (iv), (v) and, insome instances along path (vi), using any of threepossible methods available depending on the pre-dictive software and server used. The three methodsfor receiving the results of bioinformatics analysiscarried out online are as follows:-

Method 1

The search results may be viewed online at a uniquetemporary Uniform Resource Locator (URL)implemented by the software server. For example,when using the interactive results option with theBLAST program provided by the EBI server, theresearcher’s BLAST search is assigned a unique 18digit directory path to where the results of thesearch are automatically but temporarily posted,and may be viewed online as a hypertext documentwhen the analysis has completed. The search resultsare maintained online at this unique URL normallyfor a period of about 24 hours; however, some largefiles are deleted after about only 15 minutes.Regardless of how long the URL is maintainedonline, it is possible to view it, and the results, from

Figure 1. A schematic flowchart illustrating a network ofcomputers during the use of bioinformatics software online.A) The researcher’s computer on which the novel sequenceis saved. B) A local ISP, e.g. demon, AOL etc. C) A serverproviding bioinformatics software online, e.g. NCBI, Wu-BLAST. D) A server database against which the query term issearched online, e.g. EMBL

The patentability of biomolecules 121


any PC, providing the correct URL is inserted intothe web address window of a web browser.

Method 2

The search results are viewed online at a generictemporary URL generated by the software server.For example, when using the BLAST programprovided by the NCBI server, the researcher’ssequence is automatically assigned an individual 18digit request identifier (ID) number shown in an IDwindow, and then entered into a queue awaitinganalysis. Upon completion of the BLAST analysis,the researcher must then press a ‘Format’ button toview the results. Using the ID number, the serverthen automatically posts the results of the search toa temporary generic URL, linked to NCBI, whichmay be viewed online by the researcher. The serveralso provides the option of viewing the results ofdifferent BLAST searches carried out earlier usingthe NCBI server online by entering a different validrequest ID number into the ID window and viewingthese results at the same generic URL. A valid IDnumber is one which has been assigned to a recentBLAST search using NCBI, and consists of datawhich is still stored in a temporary buffer andwhich is therefore still available online. As withMethod 1, it is possible to view the URL and theresults from any PC online, providing a valid IDnumber is inserted into the ID window.

Method 3

Some servers provide the researcher with the optionof having the results delivered directly back to themvia e-mail. For example, when using the BLASTprogram provided by the IGH Montpellier server,the researcher is given the option of entering his orher private e-mail address into the relevant fieldprior to carrying out their bioinformatics analysis.Once their e-mail address has been noted, analysis iscarried out with the subsequent results being sent tothe researcher shortly thereafter.

Discussion

When assessing whether a biomolecule which hasbeen subjected to online bioinformatics analysismeets the novelty requirement for valid patentprotection, one needs to determine whether, in

doing so, that biomolecule has been ‘made availableto the public’ in terms of Art.54 EPC. Informationis said to be made ‘available’ to the public if only asingle member of the public is in a position to gainaccess to it and understand it, and if there is noobligation to maintain secrecy. If a person who wasable to gain knowledge of an invention was underan obligation to maintain secrecy, the inventioncannot be said to have been made available to thepublic, provided that person does not breach thatobligation. In addition, for a disclosure to benovelty destroying, it must be an ‘enabling dis-closure’. This means it must disclose information insufficient detail such that the invention may be putinto practice by a skilled technician. For example,in the case of a patent for a biomolecule, to benovelty-destroying, the disclosure would have togive details of the biomolecule’s sequence andsuggest at least one plausible function.

Finally, how an invention is actually madeavailable to the public is in fact immaterial. TheEPO Board of Appeal has held that the theoreticalpossibility of having access to an invention rendersit in the public domain and, therefore, publiclyavailable, whatever the means by which the inven-tion was made accessible. For example, the EPOBoard of Appeal has taken the view that if adocument in a library ‘would have been available toanyone who requested to see it’ on any particularday, then this was sufficient to establish that thedocument had been made available to the public onthat day. It is not necessary, as a matter of law, thatany member of the public would actually have beenaware that the document was available on that day,or that any member of the public had actually takennote of it. Therefore, it is not that someone wouldview an unreferenced document in a library, the factthat they could have viewed it, perhaps even bychance by walking down a random aisle and view-ing a random book on a randomly chosen shelf,may be sufficient to result in a novelty destroyingprior disclosure. Awareness, or actual inspection, ofthe document and, hence, invention, does not needto be proven in order for the disclosure to benovelty-destroying. With this, and online bioinfor-matics analysis, in mind, one needs to ascertainwhether any of the three available methods forviewing the results of such analysis (Methods 1, 2 or3), would result in those results having been ‘madeavailable to the public’.

In 2001, the World Intellectual Property Organi-sation (WIPO) Standing Committee on Patent Law

122 A. Hutter


conducted a survey on current national and regio-nal practices in Internet-related issues, in particularwith regard to publications on the Internet and thesecurity of material contained within e-mails. Therelevant document, available at (http://www.wipo.int/scp/en/documents/session_5), is SCP/5/4. The gen-eral consensus was that any website which may befreely viewed and inspected online by third partiesover the internet is regarded as a publication, andthat any information contained in a web site priorto the filing date of a patent application acts as fullycitable prior art. From the survey results, it appearsthat the relationship between the general availabil-ity/accessibility of a website and it’s effect as priorart highlighted a number of conflicting viewsbetween the countries who took part. Whilst mostcountries thought that the possibility of finding awebsite disclosing an invention using a searchengine should be taken into account, some coun-tries believed the degree of difficulty to access thecontent of any disclosure, be it online or otherwise,should not be relevant. Concerning the duration ofthe website online, the consensus was that theinformation should appear on the Internet longenough such that it could be deemed to have beenmade ‘available to the public’, this being judged ona case by case basis. On this issue, more than onecountry stated that, once the information wasposted on a website, it was a novelty-destroyingdisclosure irrespective of the length of its appear-ance on the Internet.

Hence, it appears that a publicly accessiblewebsite is analogous to a book or journal whichmay be found and read in a library, perhaps onlyfor a limited period of time, but long enough forsomeone to physically find it in the first place.Therefore, if an invention is described on a live andfreely available URL prior to the filing date of apatent application directed to that invention, andwith an enabling disclosure, then that website maywell be regarded as being a novelty-destroyingdisclosure thereby invalidating that patent applica-tion. To continue the analogy, posting the results ofa BLAST search to a URL as in either Method 1 or2 may be described as being the equivalent ofplacing a book disclosing the BLAST results in alibrary without referencing it first, which would stillrender it publicly available in terms Art.54 EPC.Theoretically, anybody could find the URL andgain access to the search results and the invention.

Using Method 1, anyone could compose a webaddress for a server hosting bioinformatics software

(e.g. http://www.ebi.ac.uk/), and a directory path(e.g. servicestmp), which may link to a page ofbioinformatics results. Each of these parts of theweb address are easily obtainable by simply explor-ing the software server’s website and determiningapproximately where results are posted. One couldthen randomly enter digits constituting the filename (e.g. 1234.html) until a valid URL showing arecent set of search results was found. It isappreciated that this may be a very time-consumingprocess since the odds of finding a valid URL couldbe as low as 1 in 1017. Moreover, because the URLshowing the results is only available online for alimited period of time, perhaps only a few hours orso, then the time during which a URL address isactually valid is relatively short. In view of thesurvey results mentioned above, the accessibility ofthe website, or the virtually ‘inaccessibility’ thereof,may be a major factor in deciding whether it can besaid to have been made available to the public ornot. Unfortunately, the absence of legal precedentsin this area makes it impossible to conclude whethera website, which is only online for a few hours andincredibly difficult to find, would amount to a priordisclosure. Nevertheless, theoretically, a valid URLcould be found, perhaps by using a powerfulnumber-generating algorithm, and it is this theore-tical possibility which may well amount to publica-tion of the URL in the eyes of the Courts. Thiswould be analogous to findings the Courts havemade in paper-based situations. Therefore, a patentapplication directed to data disclosed on an onlinewebsite as in Method 1 could be found to be invalidfor lack of novelty.

Using Method 2, anyone is able to freely visit thegeneric website hosted by the web server which hasthe ID window and in which a search ID numbermay be inserted. One could then insert ID numbersuntil a valid one was found and thereby gain accessto the URL showing a recent set of search results.As with Method 1, the likelihood of finding a validID number linked to a URL is incredibly slim.However, theoretically, it is possible and, similarly,such a method of viewing search results could beseen to be a publication and therefore novelty des-troying. However, in contrast to using Method 1,when using Method 2, an additional step has to betaken in order to view a URL showing bioinfor-matics results. Whereas in Method 1, one only hasto insert the digits directly into the web browserURL window in the hope that a valid website isfound immediately, in Method 2, one must insert



digits into the ID window and then press the‘Format’ button before being linked to a URL.Therefore, using Method 2, it is unlikely that anythird party attempting to access the results websitewill have acted in good faith, since they could notaccidentally stumble upon it by a ‘simple’ typogra-phical error.

Finally, using Method 3, the researcher’s resultsare sent from Box C to Box A in Figure 1 viae-mail. The survey conducted by the WIPO Stand-ing Committee on Patent Law also questioned thesecurity of material contained within e-mails and, ofthe 27 countries that responded on this issue, nonestated that they regarded a private e-mail asconstituting a public disclosure. This is irrespectiveof whether encryption is used, and regardless of anywarning of confidentiality which may be attached,although it does assume that the recipient is boundby a duty of confidence. The general view is that ane-mail is analogous to a letter or post-card which isconsigned to a postal delivery service. A postalemployee who sees the letter is not considered to bea member of the public, and should not use theinformation contained within the letter. Similarly,an employee of an ISP is not considered to be amember of the public and is therefore not free todisclose the information they see in an e-mail, evenif there is no bar on employees of ISPs looking atthe content of an e-mail as it passes through theirservers.

The EBI do not regard the submission of abiomolecule sequence as a query term to theirbioinformatics software as having been made pub-licly available (Peter Stoehr, personal communica-tion). The query sequence and the results aretemporarily stored by EBI, but are both deletedshortly after the search has been completed. Thesequence and search results are not inspected at allby the database and software support staff exceptfor the purpose of executing the user search andproviding user support, if necessary. Every effort ismade to keep such temporary data storage con-fidential and as secure as possible so that neither thequery sequence nor the results of the search aremade available to the public in any way. Therefore,it would appear that neither the uploading of aquery term onto the EBI software server nor therelaying of the results back to the researcher bye-mail are novelty destroying acts per se. For a thirdparty to gain access to a temporarily stored queryterm, or the results of a search, it would have to beby some form of deliberate subterfuge. Intercepting

confidential information along any of the paths (i)to (vi) in Figure 1 may be possible using today’shacking methodologies. However, it is more likelyto occur when the information is temporarily storedat either of boxes B, C or D where the sequence ofthe biomolecule consists of a single packet of datainstead of a number of separate packets of data asin paths (ii) to (v) which may be difficult tocorrectly reassemble in to SEQ ID.

Non-prejudicial disclosures

Unfortunately, things can and do go wrong. Apostal worker, sequence database employee or ISPemployee may read a letter or e-mail describing aninvention and publicly disclose what they have seen.Alternatively, a computer hacker may interceptconfidential information describing an inventionalong paths (i) to (vi) shown in Figure 1 using a‘packet sniffer’, and disclose that information tothird parties. It is evident that, in all cases, thenovelty of the invention and patent application willbe put at risk through no direct fault of thepatentee. Fortunately, if any of these circumstancesdo arise, it is possible that such actions wouldbe viewed by the Courts as a breach of confidenceby the employees or being unlawfully obtainedby a hacker, and therefore be regarded as non-prejudicial disclosures. European patent law accountsfor two scenarios in which a prior disclosure of aninvention does not prejudice the novelty of asubsequent patent application for that invention,providing the prior disclosure occurred no earlierthan six months preceding the filing date of aEuropean patent application (Art.55 EPC). Thefirst scenario is if the patentee had displayed theinvention at an official international exhibition orconference. The second scenario is if the disclosurewas due to an evident abuse in relation to thepatentee, which would include the breach ofconfidence or unlawful obtaining of informationdescribed above. Providing either of these twoscenarios occur, there is a six month period ofgrace in which a valid patent application may stillbe filed.

In order to take advantage of Art.55 EPC, thepatentee needs to show that an evident ‘abuse’ hadoccurred to their detriment. It has been suggestedthat deliberate intention to harm another party mayconstitute evident abuse, as would knowledge of thepossibility of the harm resulting from a planned

124 A. Hutter


breach of confidentiality. Therefore, one mustdetermine whether the hacker or database/ISPemployee intends to harm the patentee by carryingout an act of disclosure. It is possible that the sameconsideration may be applied to the case where anindividual intentionally attempts to access a uniqueURL showing results from Method 1 or Method 2.In this case, someone could locate a results page bythe random insertion of digits into the URL win-dow (Method 1) or ID number (Method 2). Such anact could be seen as on a par with attempting toview a website which is only accessible by a secretpassword, and therefore as an ‘abuse’ in accordancewith Art.55 EPC. Unfortunately, without thebenefit of case law to help guide practitioners, it isdifficult to say whether abuse has been committed,and each of these scenarios would have to be judgedon a case-by-case basis.

Therefore, if Art.55 EPC is to come into play, itis important to ensure that the employees or hacker,who may see the query sequence, or read the e-mail/letter by accident or intentionally, and who maysubsequently disclose the material contained in it,clearly understands that he is acting unlawfully and/or is bound by a duty of confidence. Unfortunately,the websites of EBI (and other database and soft-ware providers) do not include a clear statementindicating that they regard all information as beingconfidential. Therefore, a clear warning alerting thereader to the confidentiality of any material, be iteither a sequence uploaded onto a server, or bio-informatics results being posted on to a webpage, orbeing sent to the researcher via e-mail may increasethe chances of Art.55 EPC being applied.

Conclusion

Having examined the various methods by which theresults of online bioinformatics analysis may beviewed or returned to the researcher, it is nowpossible to categorise these alternatives according totheir risk of constituting a novelty destroyingpublication under Art.54 EPC. Of the three meth-ods discussed, it is the author’s view that having theresults returned via e-mail is likely to be the optioninvolving the least risk (Method 3). The generalconsensus is that e-mails are confidential commu-nications between a sender and a recipient and arenot viewed as public disclosures. However, not allservers provide the option of returning results bye-mail. The method posing the next least risk is

when the software server assigns a unique user IDsearch number which is used to view a URL atwhich the results are temporarily posted (Method 2).It is the author’s view that this method is less likelyto be seen as a publication because of the extra stepwhich must be taken by the researcher (or thirdparty) before activation of the results websiteonline. Using Method 1, where the results areautomatically posted onto a website, and where, intheory, anyone could potentially see the website byaccident, probably poses the highest risk in respectof acting as a publication, and it is suggested thatthis method should be avoided if at all possible.

With all three methods, if a prior disclosureoccurs as a result of abuse by third parties whohave intercepted the online or e-mailed results, thenit may still be possible to file a patent applicationwith a valid claim to novelty, providing theapplication is filed within six months of that priordisclosure, by taking advantage of Art.55 EPC. Tofacilitate this, it is suggested that online softwareservers and online databases include a clear secrecydeclaration stating that users’ query terms andresults are regarded as strictly confidential, so thatany breach of this confidence could be asserted tobe evident abuse.

It is interesting to speculate that if the Courtsadopt the view that results of bioinformaticsanalyses carried out over the web which are postedat an online website, albeit a temporary one, do actas public disclosures (Methods 1 or 2), then wecould have a rather awkward situation in which apatent directed to an online-researched biomoleculemay actually lack novelty and, therefore, be invalid.If this were to be the case, one wonders how manypatents may be invalid because they cover bio-molecules which used online bioinformatics prior tofiling. Many large bioscience research organisationsadopt the policy of maintaining ‘local’ or ‘in-house’sequence databases, and carry out all their bioinfor-matics behind a secure firewall. The advantage ofthis system is that there is no need to conductresearch online thereby avoiding any of the afore-mentioned risks associated with working over theInternet. Hence, it is the author’s view, that thismethod is the most risk-free manner in whichbioinformatics research could be conducted, andtherefore cannot be recommended strongly enoughwhere cost is not a major issue. Disadvantages ofthis system include the necessity for incredibly largehard-drives on which the sequence databases andbioinformatics software must be downloaded and



fully supported. Furthermore, it is necessary tocontinually update the sequence databases, nor-mally on a daily basis, in order to avoid thembecoming out of date.

Most smaller organisations, such as biotech start-up companies and many universities, do not havethe considerable financial reserves required to set upan in-house computing facility and manage querytool support locally. As a result, it is entirelypossible that these organisations have conducted,and are currently conducting, research over theweb, thereby disclosing their biomolecules beforethe filing date of a patent application.

Finally, if online bioinformatics is viewed asacting as a novelty destroying prior disclosure, itmay be possible to take advantage of a twelvemonth period of grace which is available for filingpatent applications in the United States. If aninvention has been disclosed anywhere in theworld, for example on a website, then a patentapplication may still be filed with a valid claim tonovelty in the US providing it is filed no later than

twelve months after that disclosure. It should benoted that this grace period may only be used whenfiling for protection in the US. When valid patentprotection is sought in countries other than the US,it is almost always necessary to ensure that aninvention has not been disclosed prior to the filingdate.

The above relates to the author’s views on thepotential risks inherent with using online bioinfor-matics, and should not be taken as legal advice forany particular circumstances. If specific advice isrequired, it is recommended that a Chartered orEuropean patent attorney be contacted.

Acknowledgement

Thanks go to the staff of Appleyard Lees, Peter Stoehr of

EMBL-EBI, Mike Barnes of GlaxoSmithKline, Chris

Southan of Gemini Genomics, and Jo Wixon, Managing

Editor of CFG for stimulating discussions and support

during the writing of this paper.

126 A. Hutter


Submit your manuscripts athttp://www.hindawi.com

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation http://www.hindawi.com

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttp://www.hindawi.com

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research


International Journal of

Microbiology

the patentability of biomolecules – does online...

Documents