covid-19 disease map, a computational knowledge … › content › 10.1101 ›...
TRANSCRIPT
-
1
COVID-19DiseaseMap,acomputationalknowledgerepositoryofSARS-CoV-2virus-hostinteractionmechanismsMarekOstaszewski1,AnnaNiarakis2,3,AlexanderMazein1,InnaKuperstein4,5,RobertPhair6,AurelioOrta-Resendiz7,8,VidishaSingh2,SaraSadatAghamiri9,MarcioLuisAcencio1,EnricoGlaab1, Andreas Ruepp10, Gisela Fobo10, Corinna Montrone10, Barbara Brauner10, GoarFrischman10,LuisCristóbalMonrazGómez4,5,JuliaSomers11,MattiHoch12,ShailendraKumarGupta12,JuliaScheel12,HannaBorlinghaus13,TobiasCzauderna14,FalkSchreiber13,14,ArnauMontagud15,Miguel Ponce de Leon15, Akira Funahashi16, YusukeHiki16, NorikoHiroi16,17,Takahiro G. Yamada16, AndreasDräger18,19,20, Alina Renz18,MuhammadNaveez12,21, ZsoltBocskei22, Francesco Messina23,24, Daniela Börnigen25, Liam Fergusson26, Marta Conti27,MariusRameil27,VanessaNakonecnij27,JakobVanhoefer27,LeonardSchmiester28,29,MuyingWang30, Emily E. Ackerman30, Jason Shoemaker30,31, Jeremy Zucker32, Kristie Oxford32,JeremyTeuton32,EbruKocakaya33,GökçeYağmurSummak33,KristinaHanspers34,MartinaKutmon35,36,SusanCoort35,LarsEijssen35,37,FriederikeEhrhart35,37,D.A.B.Rex38,DeniseSlenter35, Marvin Martens35, Robin Haw39, Bijay Jassal39, Lisa Matthews40, Marija Orlic-Milacic39,AndreaSenffRibeiro39,41,KarenRothfels39,VeronicaShamovsky42,RalfStephan39,Cristoffer Sevilla43, Thawfeek Varusai43, Jean-Marie Ravel44,45, Rupsha Fraser46, VeraOrtseifen47, Silvia Marchesi48, Piotr Gawron1,49, Ewa Smula1, Laurent Heirendt1, VenkataSatagopam1,GuanmingWu50,AndersRiutta34,MartinGolebiewski51,StuartOwen52,CaroleGoble52,XiaomingHu51,RupertW.Overall53,54,DieterMaier55,AngelaBauch55,BenjaminM.Gyori56,JohnA.Bachman56,CarlosVega1,ValentinGrouès1,MiguelVazquez15,PabloPorras43,Luana Licata57, Marta Iannuccelli57, Francesca Sacco57, Anastasia Nesterova58, AntonYuryev58, Anita de Waard59, Denes Turei60, Augustin Luna61,62, Ozgun Babur63, SylvainSoliman3, Alberto Valdeolivas60, Marina Esteban-Medina64,65, Maria Peña-Chilet64,65,66,Tomáš Helikar67, Bhanwar Lal Puniya67, Dezso Modos68,69, Agatha Treveil68,69, MartonOlbei68,69, Bertrand De Meulder70, Aurélien Dugourd60,71, Aurelien Naldi3, Vincent Noel4,Laurence Calzone4,5, Chris Sander61,62, Emek Demir11, Tamas Korcsmaros68,69, Tom C.Freeman72,FranckAugé22,JacquesS.Beckmann73,JanHasenauer28,74,OlafWolkenhauer12,EgonL.Wilighagen35,AlexanderR.Pico34,ChrisT.Evelo35,36,MarcE.Gillespie43,75,LincolnD.Stein39, Henning Hermjakob43, Peter D'Eustachio40, Julio Saez-Rodriguez60, JoaquinDopazo64,65,66,76, Alfonso Valencia77, Hiroaki Kitano78,79, Emmanuel Barillot4,5, CharlesAuffray70, Rudi Balling1, Reinhard Schneider1, and the COVID-19 Disease MapCommunity80
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
2
1. LuxembourgCentreforSystemsBiomedicine,UniversityofLuxembourg,Esch-sur-Alzette,Luxembourg2. DepartmentofBiology,Univ.Evry,UniversityofParis-Saclay,GenHotel,Genopole,91025,Evry,France 3. LifewareGroup,InriaSaclay-IledeFrance,Palaiseau91120,France4. InstitutCurie,PSLResearchUniversity,Paris,France. INSERM,U900,Paris,France.5. MINESParisTech,PSLResearchUniversity,Paris,France.6. IntegrativeBioinformatics,Inc.,346PaulAve,MountainView,CA,USA7. InstitutPasteur,HIV,InflammationandPersistenceUnit,Paris,France 8. BioSorbonneParisCité,UniversitédeParis,Paris,France9. Inserm-Institutnationaldelasantéetdelarecherchemédicale.Saint-LouisHospital1avenueClaudeVellefauxPavillon
Bazin75475Paris10. InstituteofExperimentalGenetics (IEG),HelmholtzZentrumMünchen-GermanResearchCenter forEnvironmental
Health(GmbH),IngolstädterLandstraße1,D-85764Neuherberg,Germany11. OregonHealth&SciencesUniverity;DepartmentofMolecularandMedicalGenetics;3222SWResearchDrive,Portland,
Oregon,U.S.A9723912. DepartmentofSystemsBiologyandBioinformatics,UniversityofRostock,18051Rostock,Germany13. DepartmentofComputerandInformationScience,UniversityofKonstanz,Konstanz,Germany14. Monash University, Faculty of Information Technology, Department of Human-Centred Computing,Wellington Rd,
ClaytonVIC3800,Australia15. BarcelonaSupercomputingCenter(BSC),Barcelona,Spain16. KeioUniversity,DepartmentofBiosciencesandInformatics,3-14-1HiyoshiKouhoku-kuYokohamaJapan223-852217. Sanyo-OnodaCityUniversity,FacultyofPharmaceuticalSciences,UniversitySt.1-1-1,Yamaguchi,Japan756-088418. ComputationalSystemsBiologyofInfectionsandAntimicrobial-ResistantPathogens,InstituteforBioinformaticsand
MedicalInformatics(IBMI),UniversityofTübingen,72076Tübingen,Germany19. DepartmentofComputerScience,UniversityofTübingen,72076Tübingen,Germany20. GermanCenterforInfectionResearch(DZIF),partnersiteTübingen,Germany21. RigaTechnicalUniversity,InstituteofAppliedComputerSystems,1KalkuStreet,LV-1658Riga,Latvia22. SanofiR&D,TranslationalSciences,1avPierreBrossolette91395Chilly-MazarinFrance23. DipartimentodiEpidemiologiaRicercaPre-ClinicaeDiagnosticaAvanzata,NationalInstituteforInfectiousDiseases
'LazzaroSpallanzani'I.R.C.C.S.,Rome,Italy24. COVID19INMINetworkMedicineforIDsStudyGroup,NationalInstituteforInfectiousDiseases'LazzaroSpallanzani'
I.R.C.C.S.,Rome,Italy25. BioinformaticsCoreFacility,UniversitätsklinikumHamburg-Eppendorf,Martinistraße52,20246Hamburg,Germany26. TheUniversityofEdinburgh,Royal(Dick)SchoolofVeterinaryMedicine,EasterBushCampus,Midlothian,EH259RG27. FacultyofMathematicsandNaturalSciences,UniversityofBonn,Bonn,Germany28. HelmholtzZentrumMünchen-GermanResearchCenterforEnvironmentalHealth,InstituteofComputationalBiology,
85764Neuherberg,Germany29. TechnischeUniversitätMünchen,CenterforMathematics,ChairofMathematicalModelingofBiologicalSystems,85748
Garching,Germany30. DepartmentofChemicalandPetroleumEngineering,UniversityofPittsburgh31. DepartmentofComputationalandSystemsBiology,UniversityofPittsburgh32. PacificNorthwestNationalLaboratory,902BattelleBoulevard,Richland,WA,US33. AnkaraUniversity,StemCellInstitute,CeyhunAtifKansuSt.No:16906520Cevizlidere,Ankara,Turkey34. InstituteofDataScienceandBiotechnology,GladstoneInstitutes,SanFrancisco,CA94158,US35. DepartmentofBioinformatics-BiGCaT,NUTRIM,MaastrichtUniversity,Maastricht,TheNetherlands 36. MaastrichtCentreforSystemsBiology(MaCSBio),MaastrichtUniversity,Maastricht,TheNetherlands37. MaastrichtUniversityMedicalCentre,Universiteitssingel50,6229ERMaastricht,TheNetherlands38. CenterforSystemsBiologyandMolecularMedicine,Yenepoya(DeemedtobeUniversity),Mangalore575018,India39. OntarioInstituteforCancerResearch,MaRSCentre,661UniversityAvenue,Suite510,Toronto,Ontario,CanadaM5G
0A340. NYUGrossmanSchoolofMedicine,NewYorkNY10016,US41. UniversidadeFederaldoParaná,Brasil42. NYULangoneMedicalCenter,NewYork,USA43. EMBL-EBI,MolecularSystems,WellcomeGenomeCampus,Hinxton,Cambridgeshire,CB101SD
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
3
44. UniversityofLorraine,INSERMUMR_S1256,Nutrition,Genetics,andEnvironmentalRiskExposure(NGERE),FacultyofMedicineofNancy,F-54000Nancy,France
45. Laboratoiredegénétiquemédicale,CHRUNancy,Nancy,France46. TheUniversityofEdinburgh,Queen'sMedicalResearchInstitute.47LittleFranceCrescent,Edinburgh,EH164TJ47. Senior Research Group in Genome Research of Industrial Microorganisms, Center for Biotechnology, Bielefeld
University,Universitätsstraße27,33615Bielefeld,Germany48. UppsalaUniversity,Sweden49. PoznanUniversityofTechnology,InstituteofComputingScience,ul.Piotrowo2,60-965Poznan,Poland50. DepartmentofMedical Informatics andClinicalEpidemiology,OregonHealth&ScienceUniversity,3181S.W. Sam
JacksonParkRoad,Portland,OR97239-3098,USA51. HeidelbergInstituteforTheoreticalStudies(HITS),Schloss-Wolfsbrunnenweg35,D-69118Heidelberg,Germany52. TheUniversityofManchester,DepartmentofComputerScience,OxfordRoad,Manchester,M139PL,UK53. GermanCenterforNeurodegenerativeDiseases(DZNE)Dresden,Tatzberg41,01307Dresden,Germany54. Center for Regenerative Therapies Dresden (CRTD), Technische Universität Dresden, Fetscherstraße 105, 01307
Dresden,Germany55. BiomaxInformaticsAG,Robert-Koch-Str.2,82152Planegg,Germany56. HarvardMedicalSchool,LaboratoryofSystemsPharmacology,200LongwoodAvenue,Boston,MA,US57. UniversityofRomeTorVergata,DepartmentofBiology,ViadellaRicercaScientifica1,00133Rome,Italy58. Elsevier,1600JohnFKennedyBlvd#1800,Philadelphia,PA19103,US59. Elsevier,ResearchCollaborationsUnit,71HanleyLane,Jericho,VT05465,US60. HeidelbergUnivarsity,InstituteforComputationalBiomedicine,BQ0053,ImNeuenheimerFeld267,69120Heidelberg,
Germany61. cBioCenter,DivisionsofBiostatisticsandComputationalBiology,DepartmentofDataSciences,Dana-FarberCancer
Institute,Boston,MA,02215,US 62. DepartmentofCellBiology,HarvardMedicalSchool,Boston,MA,02115,US63. UniversityofMassachusettsBoston,ComputerScienceDepartment,100WilliamT,MorrisseyBlvd,Boston,MA02125,
US64. ClinicalBioinformaticsArea,FundaciónProgresoySalud(FPS),HospitalVirgendelRocio,Sevilla,41013,Spain65. ComputationalSystemsMedicinegroup, InstituteofBiomedicineofSeville (IBIS).HospitalVirgendelRocio,Sevilla
41013,Spain66. BioinformaticsinRareDiseases(BiER).CentrodeInvestigaciónBiomédicaenReddeEnfermedadesRaras(CIBERER),
FPS,HospitalVirgendelRocío,41013,Sevilla,Spain67. UniversityofNebraska-Lincoln,DepartmentofBiochemistry,1901VineSt.,Lincoln,NE,68588,US68. QuadramInstituteBioscience,RosalindFranklinRoad,NorwichResearchPark,Norwich,NR47UQ,UK69. EarlhamInstitute,NorwichResearchPark,Norwich,NR47UZ,UK70. EuropeanInstituteforSystemsBiologyandMedicine(EISBM),Vourles,France71. Institute of Experimental Medicine and Systems Biology, Faculty of Medicine, RWTH, Aachen University, Aachen,
Germany72. TheRoslinInstitute,UniversityofEdinburghEH259RG73. UniversityofLausanne,Lausanne,Switzerland74. InterdisciplinaryResearchUnitMathematicsandLifeSciences,UniversityofBonn,Germany75. St.John’sUniversityCollegeofPharmacyandHealthSciences,Queens,NYUSA76. FPS/ELIXIR-es,HospitalVirgendelRocío,Sevilla,42013,Spain77. InstitucióCatalanadeRecercaiEstudisAvançats(ICREA),Barcelona,Spain78. SystemsBiologyInstitute,TokyoJapan79. OkinawaInstituteofScienceandTechnologyGraduateSchool80. FAIRDOMHub:https://fairdomhub.org/projects/190
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
4
AbstractWeherebydescribealarge-scalecommunityefforttobuildanopen-access,interoperable,andcomputablerepositoryofCOVID-19molecularmechanisms-theCOVID-19DiseaseMap.Wediscussthetools,platforms,andguidelinesnecessaryforthedistributeddevelopmentofitscontentsbyamulti-facetedcommunityofbiocurators,domainexperts,bioinformaticians,andcomputationalbiologists.Wehighlighttheroleofrelevantdatabasesandtextminingapproaches in enrichment and validation of the curated mechanisms. We describe thecontentsofthemapandtheirrelevancetothemolecularpathophysiologyofCOVID-19andtheanalyticalandcomputationalmodellingapproachesthatcanbeappliedtothecontentsof the COVID-19 Disease Map for mechanistic data interpretation and predictions. Weconcludebydemonstratingconcreteapplicationsofourworkthroughseveralusecases.
1.IntroductionThe coronavirus disease 2019 (COVID-19) pandemic due to severe acute respiratorysyndromecoronavirus2(SARS-CoV-2)[1]hasalreadyresultedintheinfectionofover40millionpeopleworldwide,ofwhomonemillionhavedied1.Themolecularpathophysiologythat links SARS-CoV-2 infection to the clinicalmanifestations and course of COVID-19 iscomplexand spansmultiplebiologicalpathways, cell typesandorgans [2,3].Togain theinsightsintothiscomplexnetwork,thebiomedicalresearchcommunityneedstoapproachit froma systemsperspective, collecting themechanisticknowledge scatteredacross thescientific literature and bioinformatic databases, and integrating it using formal systemsbiologystandards.
Withthisgoal inmind,we initiatedacollaborativeeffort involvingover230biocurators,domainexperts,modelersanddataanalystsfrom120institutionsin30countriestodeveloptheCOVID-19DiseaseMap,anopen-access collectionof curatedcomputationaldiagramsandmodelsofmolecularmechanismsimplicatedinthedisease[4].
To this end, we aligned the biocuration efforts of the Disease Maps Community [5,6],Reactome [7], and WikiPathways [8] and developed common guidelines utilisingstandardisedencodingandannotationschemes,basedoncommunity-developedsystemsbiology standards [9–11], and persistent identifier repositories [12]. Moreover, weintegratedrelevantknowledgefrompublicrepositories[13–16]andtextminingresources,providingameanstoupdateandrefinecontentsofthemap.ThefruitoftheseeffortswasaseriesofpathwaydiagramsdescribingkeyeventsintheCOVID-19infectiouscycleandhostresponse.
1https://covid19.who.int/
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
5
Weensured that this comprehensivediagrammaticdescriptionofdiseasemechanisms ismachine-readable and computable. This allows us to develop novel bioinformaticsworkflows,creatingexecutablenetworksforanalysisandprediction.Inthisway,themapisboth human and machine-readable, lowering the communication barrier betweenbiocurators, domain experts, and computational biologists significantly. Computationalmodelling,dataanalysis,andtheir informedinterpretationusingthecontentsof themaphave the potential to identify molecular signatures of disease predisposition anddevelopment,andtosuggestdrugrepositioningforimprovingcurrenttreatments.
COVID-19DiseaseMapisacollectionof41diagramscontaining1836interactionsbetween5499 elements, supported by 617publications andpreprints. The summary of diagramsavailableintheCOVID-19DiseaseMapcanbefoundonline2inSupplementaryMaterial1.The map is a constantly evolving resource, refined and updated by ongoing efforts ofbiocuration,sharingandanalysis.Here,wereportitscurrentstatus.
InSection2weexplainthesetupofourcommunityefforttoconstructtheinteroperablecontentoftheresource,involvingbiocurators,domainexpertsanddataanalysts.InSection3wedemonstratethatthescopeofthebiologicalmapsintheresourcereflectsthestate-of-the-artaboutthemolecularbiologyofCOVID-19.Next,weoutlineanalyticalworkflowsthatcanbeusedonthecontentsofthemap,includinginitial,preliminaryoutcomesoftwosuchworkflows,discussedindetailasusecasesinSection4.WeconcludeinSection5withanoutlooktofurtherdevelopmentoftheCOVID-19mapandtheutilityoftheentireresourceinfutureeffortstowardsbuildingandapplyingdisease-relevantcomputationalrepositories.
2https://covid.pages.uni.lu/map_contents
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
6
2.BuildingandsharingtheinteroperablecontentTheCOVID-19DiseaseMapprojectinvolvesthreemaingroups:(i)biocurators,(ii)domainexperts,and(iii)analystsandmodellers:
i. Biocurators develop a collection of systems biology diagrams focused on themolecularmechanismsofSARS-CoV-2.
ii. Domain experts refine the contents of the diagrams, supported by interactivevisualisationandannotations.
iii. Analysts andmodellers develop computationalworkflows to generatehypothesesandpredictionsaboutthemechanismsencodedinthediagrams.
All threegroupshavean importantrole in theprocessofbuilding themap,byprovidingcontent,refining it,anddefiningthedownstreamcomputationaluseof themap.Figure1illustratestheecosystemoftheCOVID-19DiseaseMapCommunity,highlightingtherolesofdifferentparticipants,available formatconversions, interoperable tools,anddownstreamuses. The information about the community members and their contributions aredisseminated via the FAIRDOMHub [17], so that content distributed across differentcollectionscanbeuniformlyreferenced.
2.1 Creating and accessing the diagrams
ThebiocuratorsoftheCOVID-19DiseaseMapdiagramsfollowtheguidelinesdevelopedbythe Community, and specific workflows of WikiPathways [8] and Reactome [7]. Thebiocurators build literature-based systems biology diagrams, representing themolecularprocesses implicated in theCOVID-19pathophysiology, their complex regulationand thephenotypic outcomes. These diagrams are main building blocks of the map, and arecomposedofbiochemicalreactionsandinteractions(furthercalledaltogetherinteractions)takingplacebetweendifferenttypesofmolecularentitiesinvariouscellularcompartments.Astherearemultipleteamsworkingonrelatedtopics,biocuratorscanprovideanexpertreviewacrosspathwaysandacrossplatforms.Thisispossible,asallplatformsofferintuitivevisualisation, interpretation, and analysis of pathway knowledge to support basic andclinicalresearch,genomeanalysis,modelling,systemsbiology,andeducation.Table1listsinformationaboutthecreatedcontent.FormoredetailsseeSupplementaryMaterial1.
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
7
Figure1:TheecosystemoftheCOVID-19DiseaseMapCommunity.ThemaingroupsofCOVID-19 Disease Map Community are biocurators, domain experts, analysts, and modellers;communicatingtorefine,interpretandapplyCOVID-19DiseaseMapdiagrams.Thesediagramsarecreated and maintained by biocurators, following pathway database workflows or standalonediagrameditors,andreviewedbydomainexperts.ThecontentissharedviapathwaydatabasesoraGitLab repository; all can be enriched by integrated resources of text mining and interactiondatabases.TheCOVID-19DiseaseMapdiagrams,availableinlayout-awaresystemsbiologyformatsand integrated with external repositories, are available in several formats allowing a range ofcomputationalanalyses,includingnetworkanalysisandBoolean,kineticormultiscalesimulations.
Both interactions and interacting entities are annotated following a uniform, persistentidentification scheme,usingeitherMIRIAMor Identifiers.org [18], and theguidelines forannotationsofcomputationalmodels[19].Viralproteininteractionsareexplicitlyannotatedwiththeir taxonomyidentifiers tohighlight findings fromstrainsother thanSARS-CoV-2.Moreover, tools like ModelPolisher [20], SBMLsqueezer [21] or MEMOTE3 help toautomaticallycomplementtheannotationsintheSBMLformatandvalidatethemodel(seealsoSupplementaryMaterial2).
3https://memote.io
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
8
Table 1. COVID-19 Disease Map contents. The table summarises biocuration resources andcontent of the map across three main parts of the repository. All diagrams are listed inSupplementaryMaterial1.
Source
Individualdiagrams Reactome WikiPathways
Diagramcontents
21diagrams1334interactions
4272molecularentities397publications
1diagram101interactions
489molecularentities227publications
19diagrams401interactions
738molecularentities61publications
Access Gitlabgit-r3lab.uni.lu/covid/models
SARS-CoVinfectionscollection
reactome.org/PathwayBrowser/#/R-HSA-9679506
COVIDpathwayscollectioncovid.wikipathways.org
Exploration TheMINERVAPlatform[22]covid19map.elixir-luxembourg.org
Guide:covid.pages.uni.lu/minerva-guide
Nativewebinterface
Guide:Linktoinstructions
Nativewebinterface
Guide:Linktoinstructions
Biocurationguidelines
Community4 Community5Platform-specific5
Community5Platform-specific6
DiagramEditors
CellDesigner7Newt8SBGN-ED[23]yEd+ySBGN9
Reactomepathwayeditor6 PathVisio[24]
Formats CellDesigner SBML [25]SBGNML[26,27]
Internal,SBMLandSBGNMLcompliant
GPML[24]
2.2 Enrichment using knowledge from databases and text mining
TheknowledgeonCOVID-19mechanismsisrapidlyevolving,asdemonstratedbytherapidgrowth of the COVID-19 Open Research Dataset (CORD-19) dataset, a source scientific
4https://docs.google.com/document/d/1DFfJZe2xjXrKMHorp_-7hlqWoNVSmx6sIzAaRzXgtQs5https://reactome.org/community/training6https://www.wikipathways.org/index.php/Help:Editing_Pathways7http://celldesigner.org8https://newteditor.org9https://github.com/sbgn/ySBGN
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
9
manuscripttextandmetadataonCOVID-19andrelatedcoronavirusresearch[28].CORD-19currentlycontainsover130,000articlesandpreprints,overfourtimesmorethanwhenitwasintroduced10.Insuchaquicklyevolvingenvironment,biocurationeffortsneedtobesupported by other repositories of structured knowledge about molecular mechanismsrelevant for COVID-19, like molecular interaction databases, or text mining resources.ContentsofsuchrepositoriesmaysuggestimprovementsintheexistingCOVID-19DiseaseMap diagrams, or establish a starting point for developing new pathways (see Section“Biocurationofdatabaseandtextminingcontent”).
InteractionandpathwaydatabasesInteractionandpathwaydatabasescontainstructuredandannotatedinformationonproteininteractionsorcausalrelationships.Whileinteractiondatabasesfocusonpairsofmolecules,offeringbroadcoverageofliterature-reportedfindings.Pathwaydatabasesprovidedetaileddescription of biochemical processes and their regulations of related interactions,supported by diagrams. Both types of resources can be a valuable input for COVID-19Disease Map biocurators, given the comparability of identifiers used for molecularannotations,andthereferencetopublicationsusedfordefininganinteractionorbuildingapathway.Table2summarisesopen-accessresourcessupportingthebiocurationofthemap.SeeSupplementaryMaterials[tools]fortheirdetaileddescription.
Table 2. Resources supporting biocuration of the COVID-19 Disease Map. They include (i)collectionsofCOVID-19interactionsbytheIMExConsortium[14]andtheSIGNOR2.0[15],(ii)non-COVIDmechanisticinteractiondatabaseOmniPath[13]and(iii)theElsevierPathwayCollection,amanuallyreconstructedopen-accessdatasetofannotatedpathwaydiagramsforCOVID-1911.
Resource Type Manuallycurated
Directed Layout COVID-19specific
IMExConsortiumdatabase[29] Interaction Yes No No Yes12[14]
SIGNOR2.0database[15] Interaction Yes Yes Yes Yes13
OmniPathdatabase[13] Interaction No Yes No No
ElsevierPathwayCollection14 Pathway Yes Yes Yes Yes9
10https://www.semanticscholar.org/cord19/download(accessedon20.10.2020)11https://data.mendeley.com/datasets/h9vs5s8fz2/draft?a=f40961bb-9798-4fd1-8025-e2a3ba47b02e12https://www.imexconsortium.org13https://signor.uniroma2.it/covid/14https://pathwaystudio.com
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
10
TextminingresourcesText-miningapproachescanhelptosievethroughsuchrapidlyexpanding literaturewithnaturallanguageprocessing(NLP)algorithmsbasedonsemanticmodelling,ontologies,andlinguisticanalysistoautomaticallyextractandannotaterelevantsentences,biomolecules,andtheirinteractions.Thisscopewasrecentlyextendedtopathwayfiguremining:decodingpathway figures into their computable representations [30].Altogether, theseautomatedworkflowsleadtotheconstructionofknowledgegraphs:semanticnetworksincorporatingontology concepts, unique biomolecule references, and their interactions extracted fromabstractsorfull-textdocuments[31].
TheCOVID-19DiseaseMapProject integrates open-access textmining resources, INDRA[32],BioKB15,AILANICOVID-1916,andPathwayStudio10.Allplatformsofferkeyword-basedsearchallowing interactiveexploration.Additionally, themapbenefits fromanextensiveprotein-proteininteractionnetwork(PPI)17generatedwithacustomtext-miningpipelineusingOpenNLP18andGNormPlus[33].ThispipelinewasappliedtotheCORD-19datasetandthe collection of MEDLINE abstracts associated with the genes in the SARS-CoV-2 PPInetwork [34] using the Entrez Gene Reference-Into-Function (GeneRIF). For detaileddescriptionsoftheresources,seeSupplementaryMaterial3.
BiocurationusingdatabaseandtextminingcontentMolecular interactionsfromdatabasesandknowledgegraphsfromtextminingresourcesdiscussedabove(fromnowoncalledaltogether‘knowledgegraphs’)haveabroadcoverageat the cost of depth of mechanistic representation. This content can be used by thebiocuratorsintheprocessofbuildingandupdatingthesystemsbiologyfocuseddiagrams.Biocuratorscanusethiscontentinthreemainways:byvisualexploration,byprogrammaticcomparison,andbydirectincorporationofthecontent.
First, the biocurators can visually explore the contents of the knowledge graphs usingavailable search interfaces to locate new knowledge and encode it in the diagrams.Moreover, solutions like COVIDminer project19, PathwayStudio andAILANI offer a visualrepresentation of a group of interactions for a better understanding of their biologicalcontext,allowingsearchbyinteractions,ratherthanjustisolatedkeywords.Finally,INDRA
15https://biokb.lcsb.uni.lu16https://ailani.ai/cgi/login_bioxm_portal.cgi17https://git-r3lab.uni.lu/covid/models/-/tree/master/Resources/Text%20mining18https://opennlp.apache.org19https://rupertoverall.net/covidminer
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
11
and AILANI offer assistant bots that respond to natural language queries and returnmeaningfulanswersextractedfromknowledgegraphs.
Second, programmatic access and reproducible exploration of the knowledge graphs ispossibleviadataendpoints:SPARQLforBioKBandApplicationProgrammingInterfacesforINDRA,AILANI,andPathwayStudio.Userscanprogrammaticallysubmitkeywordqueriesandretrievefunctions,interactions,pathways,ordrugsassociatedwithsubmittedgenelists.Thisway,otherwise time-consumingtasks likeanassessmentofcompletenessofagivendiagram,orsearchfornewliteratureevidence,canbeautomatedtoalargeextent.
Finally, biocurators candirectly incorporate the content of knowledge graphs into SBMLformat using BioKC [35]. Additionally, the contents of the Elsevier COVID-19 PathwayCollection can be translated to SBGNML20 preserving the layout of the diagrams. TheSBGNMLcontentcanthenbeconvertedintootherdiagramformatsusedbybiocurators(seeSection2.3below).
2.3 Interoperability of the diagrams and annotations
Thebiocurationof theCOVID-19DiseaseMap isdistributedacrossmultiple teams,usingvarying tools and associated systems biology representations. This requires a commonapproach to annotations of evidence, biochemical reactions,molecular entities and theirinteractions. Moreover, the interoperability of layout-aware formats is needed forcomparisonandintegrationofthediagramsinthemap.
Layout-awareformatsformolecularmechanismsTheCOVID-19DiseaseMapdiagramsareencodedinoneofthreelayout-awareformatsforstandardisedrepresentationofmolecularinteractions:SBML21[36–38],SBGNML[27],andGPML[24].TheseXML-basedformatsfocustoavaryingdegreeonuser-friendlygraphicalrepresentation,standardisedvisualisation,andsupportofcomputationalworkflows.Forthedetaileddescriptionoftheformats,seeSupplementaryMaterial1.
Each of these three languages has a different focus: SBML emphasizes standardisedrepresentation of the data model underlying molecular interactions, SBGNML providesstandardised graphical representation of molecular processes, while GPML allows for apartially standardisedrepresentationofuncertainbiologicalknowledge.Nevertheless,allthreeformatsarecenteredaroundmolecularinteractions,provideaconstrainedvocabularytoencodeelementandinteractiontypes,encodelayoutoftheirdiagramsandsupportstable
20https://github.com/golovatenkop/rnef2sbgn21here,SBMLstandsfortwoformats:CellDesignerSBMLandSBMLwithlayoutandrenderpackages
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
12
identifiers for diagram components. These shared properties, supported by a commonontology22 [39], allow cross-format mapping and enable translation of key propertiesbetweentheformats.Therefore,whendevelopingthecontentsofthemap,biocuratorsusethetoolstheyarefamiliarwith,facilitatingthisdistributedtask.
FormatinteroperabilityTheCOVID-19DiseaseMapCommunity ecosystemof tools and resources (see Figure 1)ensures interoperability between the three layout-aware formats for molecularmechanisms:SBML,SBGNML,andGPML.Essentialelementsofthissetuparetoolscapableof providing cross-format translation functionality [40,41] and supporting harmonisedvisualisation processing. Another essential translation interface is a representation ofReactomepathwaysinWikiPathwaysGPML[42]andSBML.TheSBMLexportofReactomecontenthasbeenoptimisedinthecontextofthisprojectandfacilitatesintegrationwiththeotherCOVID-19DiseaseMapsoftwarecomponents.
ThecontentsoftheCOVID-19DiseaseMapdiagramscanbedirectlytransformedintoinputsofcomputationalpipelinesanddatarepositories.BesidesthedirectuseofSBMLformatinkineticsimulations,CellDesignerSBMLfilescanbetransformedintoSBMLqual[43]usingCaSQ[44],enablingBooleanmodelling-basedsimulations(seealsoSupplementaryMaterial3). In parallel, CaSQ converts the diagrams to the SIF format23, supporting pathwaymodellingworkflowsusingsimplifiedinteractionnetworks.Notably,theGitLabrepositoryfeaturesanautomated translationof stableversionsofdiagrams intoSBMLqual.Finally,translationofthediagramsintoXGMMLformat(theeXtensibleGraphMarkupandModellingLanguage) using Cytoscape [45] or GINSim [46] allows for network analysis andinteroperabilitywithmolecularinteractionrepositories[47].
3.StructureandscopeofthemapThanks to the community effort discussed above supported by a rich bioinformaticsframework,weconstructedtheCOVID-19DiseaseMap,focussingonthemechanismsknownfrom other coronaviruses [48] and suggested by early experimental investigations[PMID:32511329]. Then, we applied the analytical and modelling workflows to thecontributed diagrams and associated interaction databases to propose initialmap-basedinsightsintoCOVID-19molecularmechanisms.
TheCOVID-19DiseaseMapisanevolvingrepositoryofpathwaysaffectedbySARS-CoV-2.Figure 2. It is currently centred on molecular processes involved in SARS-CoV-2 entry,
22http://www.ebi.ac.uk/sbo/main/23http://www.cbmc.it/fastcent/doc/SifFormat.htm
.CC-BY 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted October 28, 2020. ; https://doi.org/10.1101/2020.10.26.356014doi: bioRxiv preprint
https://doi.org/10.1101/2020.10.26.356014http://creativecommons.org/licenses/by/4.0/
-
13
replication,andhost-pathogeninteractions.Asmechanismsofhostsusceptibility,immuneresponse,cellandorganspecificityemerge,thesewillbeincorporatedintothenextversionsofthemap.
Figure2:Thestructureandcontentof theCOVID-19DiseaseMap.Theareasof focusof theCOVID-19Mapbiocuration.
The COVID-19 Map represents the mechanisms in a “host cell”. This follows literaturereportsoncellspecificityofSARS-CoV-2[3,49–53].SomepathwaysincludedintheCOVID-19Mapmaybesharedamongdifferentcelltypes,asforexampletheIFN-1pathwayfoundincellssuchasdendritic,epithelial,andalveolarmacrophages[54–58].Whileatthisstage,wedonotaddresscellspecificityexplicitlyinourdiagrams,extensiveannotationsmayallowidentificationofpathwaysrelevanttothecelltypeofinterest.
The SARS-CoV-2 infection process andCOVID-19progression follow a sequence of steps(Figure 3), starting from viral attachment and entry, which involve various dynamicprocesses on different time scales that are not captured in static representations ofpathways.Correlationofsymptomsandpotentialdrugssuggestedtodatehelpsdownstreamdataexplorationanddrugtargetinterpretationinthecontextoftherapeuticinterventions.
Other tissues/organs/systems (cardiovascular, renal, gastrointestinal, nervous)
Innate immune and inflammatory response
T cellsAdaptive immune response
Human host
ILC1, ILC-2, ILC3
Antigen-Presenting Cell
Natural killer
Systemic circulation
Renin–angiotensin–aldosterone system
(RAAS)
Host cell (Type II Pneumocyte)
Respiratory tract
Antibodies
production
Coagulation and
thrombosisCytokine release
SARS-CoV-2
B cells
Red blood cells
Th1
Th2
Macrophages
Monocytes
Viral replication cycle
Endosome and
uncoating
Genome
replication
Transcription and
translationVirion
Target cell
GranulocytesT-cell
activation
Specific T-cell
response
Alveoli
Bronchial epithelium
Nasal mucosa
COVID-19Disease Map
Vascular endothelial cell
Cellular metabolism
Viral protein interactions
Golgi
ER
CD8+
CD4+
ACE2 TMPRSS2
PAMP Signalling
Integrative stress response
Unfolded protein responseMitochondrial ETC
Viral molecules (ssRNA, dsRNA)
Mitochondria
ApoptosisJNK PathwayAutophagy
Endoplasmic
Reticulum Stress
Dendritic cells
ACE2
Viral shedding
IFN-IIL-6
IL-6
Attachment and entry
-
14
Figure3:OverviewofthemapinthecontextofCOVID-19progression.Pathwaysandcelltypesinvolved in the sequential stages of COVID-19, including some of the most common clinicalmanifestationsandmedicalmanagementfromthemomentofinfectiontothediseaseresolution,areshown.Thedistributionoftheelementsisforillustrativereferenceanddoesnotnecessarilyindicateeitheraunique/staticinterplayoftheseelementsoranunvaryingprogression.Fortheliteratureonclinicalmanifestationssee[59–65].
SupplementaryMaterial1summarisesthecontentsoftheCOVID-19DiseaseMapdiagrams,theircentralplatformofreference.TheonlineversionofthetableiscontinuouslyupdatedtoreflecttheevolvingcontentoftheCOVID-19DiseaseMap24.
3.1 Virus replication cycle and subversion of host defence
VirusattachmentandentryTransmission of SARS-CoV-2 primarily occurs through contact with respiratory drops,airborne transmission, and through contact with contaminated surfaces [66–68]. Uponcontactwiththerespiratoryepithelium,thevirusinfectscellsmostlybybindingthespikesurfaceglycoprotein(S)toangiotensin-convertingenzyme2(ACE2)withthehelpofserineprotease TMPRSS2 [69–72]. Importantly, recent results suggest viral entry using otherreceptorsof lungsand the immunesystem[73,74].Onceattached,SARS-CoV-2canentercellseitherbydirectfusionofthevirionandcellmembranesinthepresenceofproteases
24https://covid.pages.uni.lu/map_contents
Viral replication
• Attachment and entry.
• Replication cycle.• Viral proteins
interactions.
• Dendritic cells.• NK cells.• Monocytes and
macrophages.
• T cells, Th1 and Th2 response.
• B cells, antibody production.
Cytokine release and systemic inflammation
Asymptomatic/Pre-symptomatic.
Tissue damage
Target cell (respiratory tract)
Immune response and inflammation modulation
Vaccine?Pre-exposure prophylaxis?
Antivirals?
SIRS, shock.Shortness of breath.
Recovery
Coagulation
Anosmia, ageusia, cough, fever, diarrhea.
Organ damage
Multiple organ dysfunction
ARDS, complications.
Death
Host response
• Signaling.• Metabolism.
• Cellular stress.• Apoptosis.
Innate immune response
Adaptive immune response
Cellular mechanisms
Systemic response
Treatment of complications
Systemic and ventilation supportOxygen therapy
Host
RAAS
ARDS; Acute respiratory distress syndrome. RAAS; Renin-angiotensin-aldosterone system. SIRS; Systemic inflammatory response syndrome.
Interventions, potential treatments
SARS-CoV-2
Pathophysiology
Virus-host cell interactions and host response
COVID-19Disease Map
Severity and clinical manifestations
SevereMild CriticalAsymptomatic
(Lung, heart, kidney)(Nasal and respiratory epithelium, alveoli, vascular endothelial)
-
15
TMPRSS2andfurinorbyendocytosisintheirabsence.Regardlessoftheentrymechanism,the S protein has to be activated to initiate the plasma or endosomemembrane fusionprocess.While in thecellmembrane,Sprotein isactivatedbyTMPRSS2and furin, in theendosome S protein is activated by cathepsin B (CTSB) and cathepsin L (CTSL) [71,75].Activated S promotes the cell- or endosome-membrane fusion [76] with the virionmembrane,andthenthenucleocapsidisinjectedintothecytoplasm.Thesemechanismsarerepresentedinthecorrespondingdiagramsofthemap25.
ReplicationandreleaseWithinthehostcell,SARS-CoV-2hijackstheroughendoplasmicreticulum(RER)-linkedhosttranslationalmachinery.Itthensynthesisesviralproteinsreplicasepolyprotein1a(pp1a)and replicase polyprotein 1ab (pp1ab) directly from the virus (+)genomic RNA (gRNA)[48,77].Throughacomplexcascadeofproteolyticcleavages,pp1aandpp1abgiveriseto16non-structuralproteins(Nsps)[78–80].MostoftheseNspscollectivelyformthereplicationtranscription complex (RTC) that is anchored to themembraneof thedouble-membranevesicle[78,81],acoronavirusreplicationorganelleinducedbyNsps3,4,and6[82].RTCisthought to play two main roles: 1) triggering the synthesis of both (+)gRNA and(+)subgenomicmessengerRNAs(sgmRNAs)vianegative-strandedtemplates[83–85];and2) protecting intermediatedouble-strandedRNAs from the cell innate immunity sensors[86]. The (+)sgmRNAs are translated by the RER-attached translation machinery intostructural(E,M,NandS)andaccessoryproteins(Orf3a,Orf6,Orf7a/b,Orf8,Orf9b,Orf10andOrf14).Thestructuralproteinsandthenewlygenerated(+)gRNAsareassembledintonew virions in the endoplasmic reticulum-Golgi intermediate compartment. These arereleasedtotheextracellularspaceviasmooth-walledvesicles[48,77].
Endoplasmicreticulumstressandunfoldedproteinresponse
Asdiscussedabove,thevirushijackstheERtoreplicate.Productionoflargeamountsofviralproteinsexceeds theprotein foldingcapacityof theER, creatinganoverloadofunfoldedproteins.Asaresult,theunfoldedproteinresponse(UPR)pathwaysaretriggeredtoassuretheERhomeostasis,using threemainsignallingroutesofUPRviaPERK, IRE1,andATF6[87].Theirrole istomitigatethemisfoldedproteinloadandreduceoxidativestress.Theresulting protein degradation is coordinated with a decrease in protein synthesis viaeIF2alpha phosphorylation and induction of protein folding genes via the transcriptionfactorXBP1[88].WhentheERisunabletorestoreitsfunction,itcantriggercellapoptosis[89,90].
25https://covid19map.elixir-luxembourg.org/minerva/?search%3Dvirus%2520replication%2520cycle
-
16
The results are ER stress and activation of the UPR. The expression of some humancoronavirus(HCoV)proteinsduringinfection,inparticulartheSglycoprotein,mayinduceactivationoftheERstressinthehostcells[91].BasedonSARS-CoVresults,thismayleadtoactivationofthePERK[92],IRE1andinanindirectmanner,oftheATF6pathways[93].
Autophagyandproteindegradation
Processes of degrading malfunctioning proteins and damaged organelles, including theubiquitin-proteasomesystem (UPS)andautophagy [94] areessential tomaintainenergyhomeostasisandpreventcellularstress[95,96].Autophagyisalsoinvolvedincelldefence,includingdirectdestructionofthevirusesviavirophagy,presentationofviralantigens,andinhibitionofexcessiveinflammatoryreactions[97,98].
SARS-CoV-2directlyaffectstheprocessofUPS-basedproteindegradation,asindicatedbythe host-virus interactome dataset published recently [34]. This mechanism may be adefenceagainstviralproteindegradation[99].Themapdescribesindetailthenatureofthisinteraction,namelytheimpactofOrf10virusproteinontheCul2ubiquitinligasecomplexanditspotentialsubstrates.
Interactions between SARS-CoV-2 and host autophagy pathways are inferred based onresultsfromotherCoVs.AfindingthatCoVsusedouble-membranevesiclesandLC3-Iforreplication[100]maysuggestthatthevirusinducesautophagy,possiblyinATG5-dependentmanner [101], although some evidence points to the contrary [102]. Also, the CoVNsp6restricts autophagosome expansion, compromising the degradation of viral components[103].RecentlyrevealedmutationsinNsp6[104]indicateitsimportance,althoughtheexacteffectofthemutationsremainsunknown.Basedontheconnectionbetweenautophagyandthe endocytic pathway of the virus replication cycle [105], autophagy modulation wassuggested as a potential therapy strategy, either pharmacologically [96,105–107], or viafasting[108].
Apoptosis
Apoptosis,asynonymforprogrammedcelldeath,istriggeredbyvirus-hostinteractionuponinfection,astheearlydeathofthevirus-infectedcellsmaypreventviralreplication.Manyviruses block or delay cell death by expressing anti-apoptotic proteins to maximize theproduction of viral progeny [109]. In turn, apoptosis induction at the end of the viralreplication cycle might assist in viral dissemination while reducing an inflammatoryresponse.Forinstance,SARS-CoV-2[110]andMERS[111]areabletoinvokeapoptosisinlymphocytes,compromisingtheimmunesystem.
Apoptosisfollowstwomajorpathways[112],calledextrinsicandintrinsic.Extrinsicsignalsaretransmittedbydeathligandsandtheirreceptors(e.g.,FasLandTNF-alpha).Activateddeath receptors recruit adaptors like FADD and TRADD, and initiator procaspases like
-
17
caspase-8,leadingtocelldeathwiththehelpofeffectorcaspases-3and7[113,114].Inturn,theintrinsicpathwayinvolvesmitochondria-relatedmembersoftheBcl-2proteinfamily.Cellular stress causes Bcl-2 proteins-mediated release of cytochrome c from themitochondria into the cytoplasm. Cytochrome c then forms a complex with Apaf1 andrecruitsinitiatorprocaspase-9toformtheapoptosome,leadingtotheproteolyticactivationofcaspase-9.Activatedcaspase-9cannowinitiatethecaspasecascadebyactivatingeffectorcaspases 3 and 7 [114]. The intrinsic pathway is modulated by SARS-CoV molecules[115,116].Asintrinsicapoptosisinvolvesmitochondria,itsactivitymayalsobeexacerbatedbySARS-CoV-2disruptionsoftheelectrontransportchain,mitochondrialtranslation,andtransmembrane transport [34]. The resulting mitochondrial dysfunction may lead toincreasedreleaseofreactiveoxygenspeciesandpro-apoptoticfactors.
Another vital crosstalk is that of the intrinsic pathway with the PI3K-Akt pro-survivalpathway.ActivatedAkt canphosphorylate and inactivate variouspro-apoptotic proteins,includingBadandcaspase-9[117].SARS-CoVusesPI3K-Aktsignallingcascadetoenhanceinfection[118].Moreover,SARS-CoVcouldaffectapoptosis inacell-type-specificmanner[119,120].
SARS-CoVstructuralproteinsS,E,M,N,andaccessoryproteins3a,3b,6,7a,8a,and9bhavebeenshowntoactascrucialeffectorsofapoptosisinvitro.Structuralproteinsseemtoaffectmainlytheintrinsicapoptoticpathway,withp38MAPKandPI3K/Aktpathwaysregulatingcell death. Accessory proteins can induce apoptosis via different cascades and in a cell-specificmanner [114]. SARS-CoV E and 7a proteinwere shown to activate the intrinsicpathwaybyblockinganti-apoptoticBcl-XLlocalizedtotheER[121].SARS-CoVMproteinandtheionchannelactivityofEand3awereshowntointerferewithpro-survivalsignallingcascades[114,122].
3.2 Integrative stress response: endothelial damage, coagulation, and
inflammation The viral replication and the consequent immune and inflammatory responses causedamagetotheepitheliumandpulmonarycapillaryvascularendotheliumandactivatethemain intracellular defence mechanisms, as well as the humoral and cellular immuneresponses.Resultingcellularstressandtissuedamage[123,124]impairrespiratorycapacityandleadtoacuterespiratorydistresssyndrome(ARDS)[61,125,126].Hyperinflammationisaknowncomplication,causingwidespreaddamage,organfailure,anddeath,followedbyanotyetcompletelyunderstoodrapid increaseofcytokine levels(cytokinestorm)[127–129],andacuteARDS[130].Otherreportedcomplications,suchascoagulationdisturbancesand thrombosis are associated with severe cases, but the specific mechanisms are still
-
18
unknown[64,131–133],althoughsomereportssuggestthatCOVID-19coagulopathyhasadistinctprofile[134].
TheSARS-CoV-2infectiondisruptsthecoagulationcascadeandisfrequentlyassociatedwithhyperinflammation, renin-angiotensin system (RAS) imbalance and intravascularcoagulopathy [132,135–137]. Hyperinflammation leads in turn to detrimentalhypercoagulability and immunothrombosis, leading to microvascular thrombosis withfurtherorgandamage [138]. Importantly,RAS is influencedby risk factorsofdevelopingsevereformsofCOVID-19[139–141].
ACE2,usedbySARS-CoV-2forhostcellentry,isaregulatorofRASandiswidelyexpressedin the affected organs [142]. The main function of ACE2 is the conversion of AngII toangiotensin1-7(Ang1-7),andthesetwoangiotensinstriggerthecounter-regulatoryarmsofRAS [143]. The signalling via AngII and its receptor AGTR1, elevated in the infected[142,144],inducesthecoagulationcascadeleadingtomicrovascularthrombosis[145],whileAng1-7anditsreceptorMAS1attenuatetheseeffects[143].
3.4 Innate immune response
PAMPsignalling
The innate immune system detects specific pathogen-associated molecular patterns(PAMPs), through Pattern Recognition Receptors (PRRs). Detection of SARS-CoV-2 ismediatedthroughreceptorsthatrecognisedouble-strandedandsingle-strandedRNAintheendosome during endocytosis of the virus particle, or in the cytoplasm during the viralreplication. These receptorsmediate the activation of transcription factors such as AP1,NFkappaB, IRF3, and IRF7, responsible for the transcription of antiviral proteins, inparticular,interferon-alphaandbeta[146,147].
SARS-CoV-2reducestheproductionoftypeIinterferonstoevadetheimmuneresponse[49].Thedetailedmechanism isnotclearyet;however,SARS-CoVMprotein inhibits the IRF3activation[148]andsuppressesNFkappaBandCOX2transcription.Atthesametime,SARS-CoVNproteinactivatesNFkappaB[149],sotheoverallimpactisunclear.ThesepathwaysarealsonegativelyregulatedbySARS-CoVnsp3papain-likeproteasedomain(PLPro)[150].
Themapcontainstheinitialrecognitionprocessoftheviralparticlebytheinnateimmunesystemandtheviralmechanismstoevadetheimmuneresponse.Itprovidestheconnectionbetweenvirusentry(detectingtheviralendosomalpatterns),itsreplicationcycle(detectioncytoplasmic viral patterns), and the effector pathways of pro-inflammatory cytokines,especiallyoftheinterferontypeIclass.ThelatterseemstoplayacrucialbutcomplexroleinCOVID-19pathology:bothnegative[151,152]andpositiveeffects[3,153]ofinterferonsonvirusreplicationhavebeenreported.
-
19
InterferontypeIsignalling
Interferons(IFNs)arecentralplayersintheantiviralimmuneresponseofthehostcell[55],specifically affected by SARS-CoV-2 [154–157]. Type I IFNs are induced upon viralrecognitionofPAMPsbyvarioushostPRRs [48]asdiscussedearlier.The IFN-Ipathwaydiagram represents the activation of TLR7 and IFNAR and the subsequent recruiting ofadaptor proteins and the downstream signalling cascades regulating key transcriptionfactorsincludingIRF3/7,NF-kappaB,AP-1,andISRE[48,158].Further,themapshowsIRF3-mediatedinductionofIFN-I,affectedbytheSARS-CoV-2proteins.SARS-CoVNsp3andOrf6interferewithIRF3signalling[159,160]andSARS-CoVM,N,Nsp1andNsp3actasinterferonantagonists[48,150,158,161,162].Moreover,coronavirusesORF3a,ORF6andnsp1proteinscan repress interferon expression and stimulate the degradation of IFNAR1 and STAT1duringtheUnfoldedProteinResponse(UPR)[163,164].
AnothermechanismofviralRNArecognitionisRIG-likereceptorsignalling[58],leadingtoSTING activation [165], and via the recruitment of TRAF3, TBK1 and IKKepsilon tophosphorylationofIRF3[56].ThisinturninducesthetranscriptionofIFNsalpha,betaandlambda[166].SARS-CoVviralpapain-like-proteases,containedwithinthensp3andnsp16proteins,inhibitSTINGandthedownstreamIFNsecretion[167].Inlinewiththishypothesis,SARS-CoV-2 infectionresults inaunique inflammatoryresponsedefinedby low levelsofIFN-Iandhighexpressionofcytokines[58,168].TheIFNlambdadiagramdescribestheIFNLreceptor signaling cascade [169], including JAK-STAT signaling and the induction ofInterferon Stimulated Genes, which encode antiviral proteins [170]. The interactions ofSARS-CoV-2proteinswiththeIFNLpathwayarebasedontheliterature[171]orSARS-CoVhomology[58].
Alteredhostmetabolism
Metabolicpathwaysgoverntheimmunemicroenvironmentbymodulatingtheavailabilityofnutrientsandcriticalmetabolites[172].Infectiousentitiesreprogramhostmetabolismtocreatefavourableconditionsfortheirreproduction[173].SARS-CoV-2proteinsinteractwithavarietyofimmunometabolicpathways,severalofwhicharedescribedbelow.
Hemecatabolismisawell-knownanti-inflammatorysysteminthecontextofinfectiousandautoimmune diseases [174,175]. The main effector of this pathway, heme oxygenase-1(HMOX1) was found to interact with SARS-CoV-2 Orf3a, although the nature of thisinteraction remains ambiguous [34,176]. HMOX1 cleaves heme into carbon monoxide,biliverdin (then reduced to bilirubin), and ferrous iron [PMID:31396090]. Biliverdin,bilirubin,andcarbonmonoxidepossesscytoprotectiveproperties,andhaveshownpromiseas immunomodulatory therapeutics [177,178]. Importantly, activation of HMOX1 alsoinhibits the NLRP3 inflammasome [178–180], which is a pro-inflammatory andprothromboticmultiproteinsystem[181]highlyactiveinCOVID-19[182–184].Itmediates
-
20
production of the pro-inflammatory cytokines IL-1B and IL-18 via caspase-1 [181]. TheSARS-CoVOrf3a,E,andOrf8aincitetheNLRP3inflammasome[185–188].Still,thepotentialoftheHMOX1pathwaytofightCOVID-19inflammationremainstobetested[176,189,190]despitepromisingresultsinothermodelsofinflammation[176,178,191–193].
The tryptophan-kynurenine pathway is closely related to heme metabolism. The rate-limitingstepofthispathwayiscatalysedbytheindoleamine2,3dioxygenaseenzymes(IDO1andIDO2)indendriticcells,macrophages,andepithelialcellsinresponsetoinflammatorycytokineslikeIFN-gamma,IFN-1,TGF-beta,TNF-alpha,andIL-6[194–196].CrosstalkwiththeHMOX1pathwayalsoincreasestheexpressionofIDO1andHMOX1inafeed-forwardmanner. Metabolomics analyses from severe COVID-19 patients revealed enrichment ofkynurenines and depletion of tryptophan, indicating robust activation of IDO enzymes[197,198].Depletion of tryptophan [173,199,200] and kynurenines and their derivativesaffecttheproliferationandimmuneresponseofarangeofTcells[176,201–205].However,despitehighlevelsofkynureninesinCOVID-19,CD8+T-cellsandTh17cellsareenrichedinlungtissue,andT-regulatorycellsarediminished[206].ThisraisesthequestionofwhetherandhowtheimmuneresponseelicitedinCOVID-19evadessuppressionbythekynureninepathway.
The SARS-CoV-2 protein Nsp14 interacts with three human proteins: GLA, SIRT5, andIMPDH2 [34]. The galactose metabolism pathway, including the GLA enzyme [207], isinterconnected with amino sugar and nucleotide sugar metabolism. SIRT5 is a NAD-dependent desuccinylase and demalonylase regulating serine catabolism, oxidativemetabolism and apoptosis initiation [208–210]. Moreover, nicotinamide metabolismregulated by SIRT5 occurs downstream of the tryptophan metabolism, linking it to thepathways discussed above. Finally, IMPDH2 is the rate-limiting enzyme in the de novosynthesis of GTP, allowing regulation of purine metabolism and downstream potentialantiviraltargets[211,212].
Thepyrimidinesynthesispathway, tightly linkedtopurinemetabolism,affectsviralDNAandRNAsynthesis.Pyrimidinedeprivationisahosttargetedantiviraldefencemechanism,whichblocksviralreplicationininfectedcellsandcanberegulatedpharmacologically[213–215]. It appears that componentsof theDNAdamage response connect the inhibitionofpyrimidinebiosynthesistotheinterferonsignallingpathway,probablyviaSTING-inducedTBK1 activation that amplifies interferon response to viral infection, discussed above.InhibitionofdenovopyrimidinesynthesismayhavebeneficialeffectsontherecoveryfromCOVID-19[215];however,thismayhappenonlyinasmallgroupofpatients.
-
21
3.5 Biocuration roadmap
COVID-19pathwaysfeaturedintheprevioussectioncovermechanismsreportedsofar.Still,certainaspectsof thediseasewerenotrepresented indetailbecauseof theircomplexity,namelycell-type-specific immuneresponse,andsusceptibilityfeatures.Theirmechanisticdescription isofgreat importance,assuggestedbyclinicalreportsonthe involvementofthesepathwaysinthemolecularpathophysiologyofthedisease.Themechanismsoutlinedbelowwillbethenexttargetsinourcurationroadmap.
Celltype-specificimmuneresponseCOVID-19causesseriousdisbalanceinmultiplepopulationsofimmunecells.Somestudiesreport that COVID-19 patients have a significant decrease of peripheral CD4+ and CD8+cytotoxicTlymphocytes(CTLs),Bcells,NKcells,aswellashigherlevelsofabroadrangeofcytokinesandchemokines[128,216–219].ThediseasecausesfunctionalexhaustionofCD8+CTLs andNK cells, inducedby SARS-CoV-2 Sprotein andby excessivepro-inflammatorycytokineresponse[217,220].Moreover,theratioofnaïve-to-memoryhelperT-cells,aswellas thedecreaseofTregulatorycells,correlatewithCOVID-19severity[206].Conversely,high levelsofTh17and cytotoxicCD8+T-cellshavebeen found in the lung tissue [221].Pulmonaryrecruitmentoflymphocytesintotheairwaysmayexplainthelymphopeniaandtheincreasedneutrophil-lymphocyteratioinperipheralbloodfoundinCOVID-19patients[216,222,223].Inthisregard,anabnormalincreaseoftheTh17:Tregcellratiomaypromotethe release of pro-inflammatory cytokines and chemokines, increasing disease severity[224].
SusceptibilityfeaturesofthehostSARS-CoV-2 infection isassociatedwith increasedmorbidityandmortality in individualswithunderlyingchronicdiseasesoracompromisedimmunesystem[225–228].Groupsofincreased risk are men, pregnant and postpartum women, and individuals with highoccupationalviralexposure[229–231].OthersusceptibilityfactorsincludetheABObloodgroups[232–240]andrespiratoryconditions[241–246].
Importantly,age isoneof thekeyaspectscontributingtotheseverityof thedisease.Theelderly are at high risk of developing severe or critical disease [227,247]. Age-relatedelevated levels of pro-inflammatory cytokines (inflammation) [247–250],immunosenescenceandcellularstressofageingcells[125,227,247,251,252]maycontributetotherisk.Incontrast,childrenaregenerallylesslikelytodevelopseveredisease[253,254],withtheexceptionofinfants[125,255–257].However,somepreviouslyhealthychildrenandadolescents can develop a multisystem inflammatory syndrome following SARS-CoV-2infection[258–262].
-
22
Several genetic factorshavebeenproposedand identified to influence susceptibility andseverity,includingtheACE2gene,HLAlocus,errorsinfluencingtypeIIFNproduction,TLRpathways,myeloidcompartments,aswellascytokinepolymorphisms[156,235,263–269].
Weaimtoconnectthesusceptibilityfeaturestospecificmolecularmechanismsandbetterunderstand the contributing factors. This can lead to a series of testable hypotheses,including the roleof vitaminD counteractingpro-inflammatory cytokine secretion [270–272] in an age-dependentmanner [247,273], andmodifying the severity of the disease.Anotherexampleofa testablehypothesismaybe that the immunephenotypeassociatedwithasthmainhibitspro-inflammatorycytokineproductionandmodifiesgeneexpressionintheairwayepithelium,protectingagainstsevereCOVID-19[245,246,274].
4.BioinformaticsanalysisandcomputationalmodellingroadmapforhypothesisgenerationIn order to understand complex and often indirect dependencies between differentpathways and molecules, we need to combine computational and data-driven analyses.Standardised representation and programmatic access to the contents of the COVID-19DiseaseMapenablethedevelopmentofreproducibleanalyticalandmodellingworkflows.Here,wediscuss the range of possible approaches anddemonstrate preliminary results,focusingoninteroperability,reproducibility,andapplicabilityofthemethodsandtools.
Our goal is to work on the computational challenges as a community, involving thebiocuratorsanddomainexperts intheanalysisoftheCOVID-19DiseaseMapandrelyontheirfeedbacktoevaluatetheoutcomes.Inthisway,weaimtoidentifyapproachestotacklethecomplexityandthesizeofthemap,proposingastate-of-the-artframeworkforrobustanalysis,reliablemodels,andusefulpredictions.
4.1 Data integration and network analysis
Visualisationofomicsdatacanhelpcontextualisethemapwithexperimentaldatacreatingdata-specificblueprints.Theseblueprintscouldbeusedtohighlightpartsofthemapthatareactiveinoneconditionversusanother(treatmentversuscontrol,patientversushealthy,normal versus infected cell, etc.). Combining information contained in multiple omicsplatformscanmakepatientstratificationmorepowerful,byreducingthenumberofsamplesneededorbyaugmenting theprecisionof thepatientgroups [275,276].Approaches thatintegratemultipledatatypeswithouttheaccompanyingmechanisticdiagrams[277–279]producepatientgroupingsthataredifficulttointerpret.Inturn,classicalpathwayanalysesoftenproducelonglistsmixinggenericandcell-specificpathways,makingitchallengingto
-
23
pinpoint relevant information. Using disease maps to interpret omics-based clustersaddressestheissuesrelatedtocontextualisedvisualdataanalytics.
FootprintbasedanalysisFootprintsaresignaturesofamolecularregulatordeterminedbytheexpressionlevelsofitstargets[280].Forexample,afootprintcancontaintargetsofatranscriptionfactor(TF)orpeptides phosphorylated by a kinase. Combining multiple omics readouts and multiplemeasurements can increase the robustnessof such signatures.Nevertheless, anessentialcomponent is the mechanistic description of the targets of a given regulator, allowingcomputation of its footprint. With available SARS-CoV-2 related omics and interactiondatasets[281],itispossibletoinferwhichTFsandsignallingpathwaysareaffecteduponinfection[282].CombiningtheCOVID-19Diseasemapregulatoryinteractionswithcuratedcollections of TF-target interactions like DoRothEA [283] will provide a contextualisedevaluationoftheeffectofSARS-CoV-2infectionattheTFlevel.
Viral–hostinteractomeThevirus–hostinteractomeisanetworkofvirus-humanprotein-proteininteractions(PPIs)thatcanhelpunderstandingthemechanismsofdisease[34,284–286].Itcanbeexpandedbymergingvirus-hostPPIdatawithhumanPPIandproteindata[287]todiscoverclustersofinteractionsindicatinghumanmechanismsandpathwaysaffectedbythevirus[288].TheseclustersfirstofallcanbeinterpretedatthemechanisticlevelbyvisualexplorationofCOVID-19 Disease Map diagrams. In addition, these clusters can potentially reveal additionalpathways to add to the COVID-19 Disease Map (e.g., E protein interactions or TGFBetadiagrams)orsuggestnewinteractionstointroduceintotheexistingdiagrams.
4.2 Mechanistic and dynamic computational modelling
Computationalmodellingisapowerfulapproachthatenablesinsilicoexperiments,producestestablehypotheses,helpselucidateregulationand,finally,cansuggestviapredictionsnoveltherapeutictargetsandcandidatesfordrugrepurposing.
MechanisticpathwaymodellingMechanisticmodelsofpathwaysallowbridgingvariationsatthescaleofmolecularactivitytovariationsatthelevelofcellbehaviour.Thiscanbeachievedbycouplingthemolecularinteractions of a given pathway with its endpoint and by contextualising the molecularactivity using omics datasets. HiPathia is such a method, processing transcriptomic orgenomicdatatoestimatethefunctionalprofilesofapathwayconditionedbythedatastudiedand linkable to phenotypes such as disease symptoms or other endpoints of interest[289,290]. Moreover, such mechanistic modelling can be used to predict the effect of
-
24
interventionsas,forexample,theeffectoftargeteddrugs[291].HiPathiaintegratesdirectlywiththediagramsoftheCOVID-19MapusingtheSIFformatprovidedbyCaSQ(seeSection2.3),aswellaswiththeassociatedinteractiondatabases(seeSection2.2).
Thedrawbackofapproaches likeHiPathia is theircomputationalcomplexity, limitingthesize of the diagrams they can process. An approach to large-scale mechanistic pathwaymodellingistotransformthemintocausalnetworks.CARNIVAL[292]combinesthecausalrepresentation of networks [13] with transcriptomics, phosphoproteomics, ormetabolomics data [280] to contextualise cellular networks and extract mechanistichypotheses. The algorithm identifies a set of coherent causal links connecting upstreamdrivers suchas stimulationsormutations todownstreamchanges in transcription factoractivities.
DiscretecomputationalmodellingAnalysisofthedynamicsofmolecularnetworksisnecessarytounderstandtheirdynamicsanddeepenourunderstandingofcrucialregulatorsbehinddisease-relatedpathophysiology.Discretemodelling frameworkprovidesthispossibility.COVID-19DiseaseMapdiagrams,translated to SBML qual (see Section 2.3), can be directly imported by tools like CellCollective[293]orGINsim[46]foranalysis.Preservingannotationsandlayoutinformationensurestransparencyandreusabilityofthemodels.
Importantly, Cell Collective is an online user-friendlymodelling platform26 that providesfeaturesforreal-timeinsilicosimulationsandanalysisofcomplexsignallingnetworks.Theplatformallowsuserswithoutcomputationalbackgroundtosimulateoranalysemodelstogenerate and prioritise new hypotheses. References and layout are used for modelvisualisation,supportingtheinterpretationoftheresults.Themathematicsandcodebehindeachmodel,however,remainaccessibletoallusers. Inturn,GINsimisa toolprovidingawiderangeofanalysismethods,includingefficientidentificationofthestatesofconvergenceofagivenmodel(attractors).Modelreductionfunctionalitycanalsobeemployedtofacilitatetheanalysisoflarge-scalemodels.
Multiscaleandstochasticcomputationalmodelling
Viralinfectionandimmuneresponsearecomplexprocessesthatspanmanydifferentscales,frommolecular interactions tomulticellular behaviour. Themodelling and simulation ofsuchcomplexscenariosrequireadedicatedmultiscalecomputationalarchitecture,wheremultiplemodelsruninparallelandcommunicateamongthemtocapturecellularbehaviourandintercellularcommunications.Multiscaleagent-basedmodelssimulateprocessestaking
26https://cellcollective.org
-
25
placeatdifferenttimescales,e.g.,diffusion,cellmechanics,cellcycle,orsignaltransduction[294], proposed also for COVID-19 [295]. PhysiBoSS [296] allows such simulation ofintracellularprocessesbycombiningthecomputationalframeworkofPhysiCell[297]withMaBoSS[298]toolforstochasticsimulationoflogicalmodelstostudyoftransienteffectsand perturbations [299]. Implementation of detailed COVID-19 signalling models in thePysiBoSS framework may help to better understand complex dynamics of multi-scaleprocessesasinteractionsandcrosstalkbetweenimmunesystemcomponentsandthehostcellinCOVID-19.
4.3 Case study: RNA-Seq-based analysis of transcription factor (TF) activity
In this case study, we combine computational approaches discussed above and presentresults derived from omics data analysis on the COVID-19 Disease Maps diagrams. WemeasuredtheeffectofCOVID-19atthetranscriptionfactor(TF)activitylevelbyapplyingVIPER[300]combinedwithDoRothEAregulons[283]onRNA-seqdatasetsofthe SARS-CoV-2infectedcellline[168].Then,wemappedtheTFsnormalisedenrichmentscore(NES)ontheInterferontypeIsignallingpathwaydiagramoftheCOVID-19DiseaseMapusingtheSIFfilesgeneratedbyCaSQ(seeSection2.3).AshighlightedinFigure4,ourmanuallycuratedpathwayincludedsomeofthemostactiveTFsafterSARS-CoV-2infection,suchasSTAT1,STAT2,IRF9andNFKB1.Thesegenesarewellknowntobeinvolvedincytokinesignallingandfirstantiviralresponse[301,302].Interestingly,theyarelocateddownstreamofvariousviralproteins (E,S,Nsp1,Orf7aandOrf3a)andmembersof theMAPKpathway (MAPK8,MAPK14andMAP3K7).SARS-CoV-2infectionisknowntopromoteMAPKactivation,whichmediatesthecellularresponsetopathogenicinfectionandpromotestheproductionofpro-inflammatorycytokines[281].Altogether,weidentifiedsignalingeventsthatmaycapturethemechanisticresponseofthehumancellstotheviralinfection.
-
26
Figure 4: the Interferon type I signalling pathway diagram of the COVID-19 Disease MapintegratedwithTFactivityderivedfromtranscriptomicsdataafterSARS-CoV-2infection.AzoomwasappliedintheareacontainingthemostactiveTFs(rednodes)afterinfection.Nodeshapes:host genes (rectangles), host molecular complex (octagons), viral proteins (V shape), drugs(diamonds)andphenotypes(triangles).
4.4 Case study: RNA-seq-based analysis of pathway signalling
Inthisusecase,theHipathia[289]algorithmwasusedtocalculatethelevelofactivityofthesubpathways from the COVID-19 Apoptosis diagram, with the aim to evaluate whetherCOVID-19DiseaseMapdiagramscanbeusedforpathwaymodellingapproach.Tothisend,apublicRNA-seqdatasetfromhumanSARS-CoV-2infectedlungcells(GEOGSE147507)wasused.First,theRNA-seqgeneexpressiondatawasnormalizedwiththeTrimmedmeanofM
-
27
values (TMM)normalization [303], thenrescaled torange [0;1] for thecalculationof thesignalandnormalisedusingquantilenormalisation[304].Thenormalisedgeneexpressionvalueswereusedtocalculatethelevelofactivationofthesubpathways,thenacase/controlcontrastwithaWilcoxontestwasusedtoassessdifferencesinsignalingactivitybetweenthetwoconditions.
Figure5.RepresentationoftheactivationlevelofApoptosispathwayinSARS-CoV-2infectedlung cell lines. The activation levels have been calculated using transcriptional data fromGSE147507 and Hipathiamechanistic pathway analysis algorithm. Each node represents a gene(ellipse),ametabolite(circle)ora function(square).Thepathwayiscomposedofcircuits fromareceptorgene/metabolitetoaneffectorgene/function,withinteractionssimplifiedtoinhibitionsoractivations(seeSection2.3,SIFformat).Significantlyderegulatedcircuitsarehighlightedbycolorarrows(red:activatedininfectedcells).Thecolorofthenodecorrespondstothelevelofdifferentialexpressionof eachnode in SARS-CoV-2 infected cells vsnormal lung cells.Blue: down-regulatedelements,red:up-regulatedelements,white:elementswithnotstatisticallysignificantdifferentialexpression.Hipathiacalculatestheoverallcircuitactivation,andcanindicatederegulatedinteractionevenifinteractingelementsarenotindividuallydifferentiallyexpressed.
Results of the Apoptosis pathway analysis can be seen in Figure 5 and SupplementaryMaterial 5. Importantly, Hipathia calculates the overall activation of circuits (series ofcausally connected elements), and can indicatederegulated interactions resulting fromacumulativeeffect,evenifinteractingelementsarenotindividuallydifferentiallyexpressed.Whendiscussingdifferentialactivation,werefertothecircuits,whileindividualelements
-
28
arementionedasdifferentiallyexpressed.Theanalysisshowsanoveractivationofseveralcircuits,specificallytheoneendingintheeffectorproteinBAX.ThisoveractivationseemstobeledbytheoverexpressionoftheBADprotein, inhibitingBCL2-MCL1-BCL2L1complex,which in turn inhibits BAX. Indeed, SARS-CoV-2 infection can invoke caspase8-inducedapoptosis[305],whereBAXtogetherwiththeripoptosome/caspase-8complex,mayactasa pro-inflammatory checkpoint [306]. This result is supported by studies in SARS-CoV,showing BAX overexpression following in