Six Laws of Open Source Drug Discovery Matthew H

DOI:10.1002/cmdc.201900565 Editorial Six Laws of Open Source Drug Discovery Matthew H. Todd[a] 1. Open Source Applied to Drug Discovery ble, but not the transformative feature of initiatives like Wikipe- dia. In open source the community participates in peer-re- Creating and developing new medicinesrequires astream of viewedpublicwork. Depending on the exact licence of the innovations. Many of these are in the technical side of our dis- project, one will have high levels of freedomtoact. Crucially, cipline:methods in organic synthesis, assay technologies, ma- one may act while the work is happening, rather than merely chine learning, or approaches based in fundamentalbiology. after it is complete. Youare aplayer,not an observer.There is Some innovations arise in allied disciplines of economics, or cooperation towards goals that operates alongside an open law.But in parallel we must try to notice when our work pat- competition in how work is done, i.e.,collaborationwithin a terns are constrainedbysocial structures that may limit our ratherbrutal, open arena of expertise. abilitiestofunctionmost effectively.Are we going about Over the last 15 years, Ihave worked with large numbers of things in the right way?Are we innovating in the way that we people who are interested in open source in drug discovery work? because the approach promises to be able to solve problems There are large-scale initiatives in pharma trying to address in exactly the way that the pharma industry,ifacting alone, this under the banner of “open innovation”.[1] The term is neb- cannot. If one values competition, as Isuspectall scientists do, ulous, but there is typically some re-orientation of acompany then acompetition of approaches must be seen as healthy.If to be more outward-facing—to place problemsinthe public open source can competewith proprietary methods in acom- domain and attempt to broaden the net of expertise in order mercial space as valuableastelecommunications(i.e.,the proj- to solve those problems faster.The broad range of such initia- ects leading to Android vs. iOS) can the same competition of tives means that while some are highly valuable, others are ideas benefitanother commercially-important area:drug dis- branding exercises. Either way they will in general not change covery? the way we work:just as for secretive, inward-facing endeav- Istrongly suspectitcan. The answer depends on the dis- ours, participants in open innovationprojects typically work in ease. But Iamrunningthis experiment with others because closed groups, revealing only that which is felt worth sharing. the answer is not yet clear. The open part of open innovation is the shared problem, but This article is intended to clarify key terms. One cannot set not necessarily the proposed, and attempted, solutions. off for the moon without some basic principles of operation, Examples of an alternative way of workingare all around us, and vaguenesscan be worse than nothing. We ought to be but largely viewed as insufficiently seriousfor science. Wikipe- clear what “open source drug discovery” means. When Istart- dia has transformed the way knowledge is curated,aggregat- ed in this endeavour the term was aspirational,but we have ed, and shared, yet is probably the tool that scientists deny now run several successful projectswith aclear and unified set using mostfrequently.The softwarethat underpins the inter- of principles, and Iwould like to highlight these because they net was built using amethod similartothat behind Wikipedia, are useful. Operations have been distilled to six laws that have specifically that the details can be seen by anyone andpeople held stronginthe real world and which are the subjectofthis can work together openly to improvethe content or take it in article. anew direction. There is no appetite for secrecyorsilos because those stifle our ability to access the best ideas, and why 2. How the Laws Came About would you want to do that?This methodisopen source,dis- tinct from open innovation and to be distinguished from open access,which term refers to our ability to read papers—valua- Aswith any endeavourdesigned to elicit maximum inclusion and productivity,there is afine line to navigate between free- dom to operate on the one hand and asense of order and uni- [a] Prof. M. H. Todd fied purpose on the other.Yuval Noah Harari makes anice SchoolofPharmacy University College London point in his book Sapiens that corporationsexist beyondthe 29–39 Brunswick Square, WC1N 1AX London (UK) physical–that they may possess alasting identitywithoutany The ORCID identification number(s) for the author(s) of this articlecan be permanent assets.[2] It is the same with open source initiatives. found under: The lack of atangible “thing” or “building” remains one of the https://doi.org/10.1002/cmdc.201900565. most challenging ideas for those coming to an open source 2019 The Author. PublishedbyWiley-VCH Verlag GmbH &Co. KGaA. project for the first time, yet the fluidityofits structure and This is an openaccessarticleunder the termsofthe Creative Commons AttributionLicense, which permits use, distribution and reproduction in constitution provides the kind of resilience you observe when any medium, provided the original work is properly cited. trying to spear ashoal of fish. “What is an open source proj- ChemMedChem 2019, 14,1804 –1809 1804 2019 The Authors. PublishedbyWiley-VCH Verlag GmbH &Co. KGaA, Weinheim Editorial ect?” is acomplicated question, but ashort form answer to date, over 300 people on the four OSM campaigns. The might be “people, andtheir enthusiasm for ashared public Laws can apply equally well to the next campaigns that I, or mission”. others,create under OSM’sbanner or to similarprojects in The SixLaws attempt to capture someofthis structure, but other areas of drug discovery that people might want to run.[8] were not set up apriori. They arose from the first project Iran The Laws are intended to guide behaviour,tofree people to in an open way:the search for arobustroute to an enantio- act to the best of their ability within aframework that pro- pure version of the world’smost widely used anthelmintic, pra- motesadistinctive discoveryprocess. They do not hasp and ziquantel.[3] The need for such an improvement had been ex- hoop the contributors, and can be changed in the future if plicitly mentionedasaresearch priority by the tropical disease they are found to be faulty. research divisionofthe World Health Organisation. To solve it, The first three Laws clarify day-to-day operations. The others Iproposedsome chemically elaborate approaches and placed are more subtle, big-picture concepts. them online in an attempt to kick-start an open approach to solving this chiral switch. My intention was to mimic the hive mind that was, in late 2004, having amajor impact on the con- 3. The Laws struction of new software and Wikipedia. The project became busy once we were modestly funded. At this point reality col- lided with the grandeurofgrant proposals and the scientific approachwewere takingchanged in response to precisely the expert collective intelligence Ihad hoped for,freely given and mostly from the private sector.The solution emerged swiftly and was solid. The science was unquestionably accelerated through the inputs of well-qualified strangers. Over adinnerinCape Town hostedbyKellyChibaleIgot talking with TimWells, the CSO of the Medicines for Malaria Venture (MMV), who asked the obviousquestion:could we use the same approach to the discovery not of betterchemical routes, but of new chemicalentities? The conversation went back and forth until we decided that conversation can only take people so far,and it is sometimesbetter to learn by doing. Open SourceMalaria (OSM, initially called Open Source Drug Discovery for Malaria)was born from that conversation and the invaluable wisdom of the MMV team of Jeremy Bur- rows and Paul Willis. Irealized that this might turn into a larger endeavour:there were no such initiatives anywhere else. Genomics collaborations, such as the Human Genome Project,had pioneered the sharingofdata sets with clear com- mercial potential, downstream. There continues to be agreat deal of impactful research into open source tools in cheminfor- matics.[4] The Open SourceDrug Discovery project in India was, despite its name, operating acrowdsourcinginitiative as op- posed to something that was open source.[5] The StructuralGe- nomicsConsortium were pioneering the bold sharingofchem- 1) All Data and Ideas Are Freely Shared. ical probes (not drugs or their analogs)and werenot proposing the sharing of the full research cycle that led to them.[6] We This is adeceptively short Law that we oughttounpick if we were proposing something different—secrecy-free creationof are not to miss some essential features. We might callit“The new medicines. Condensed” law.When experiments are performed, those To minimize confusion, it was important to capture the core need to be recorded in alaboratory notebook (obviously), and principles of OSM in away that would be simple to under- that notebook(wherever it is) needs to be available to read in stand for potentialincomersand to ensure those joining in its entirety.This means it needs to be online and, to be of any knew the level of mischief they would be getting into. So on practical use, not behind apassword wall. It is remarkable how July 25th 2011, led by what Ifelt to be the most important les- many academic scientists are still using paper lab books that sons learned from the WHO project, Iwrote down the

Load more