D480–D487 Nucleic Acids Research, 2022, Vol. 50, Database issue Published online 25 November 2021 https://doi.org/10.1093/nar/gkab1082 DisProt in 2022: improved quality and accessibility of protein intrinsic disorder annotation Federica Quaglia 1,2 , Bálint Mészáros 3 , Edoardo Salladini 2 , András Hatos 2 , Rita Pancsa 4 , Lucı́a B. Chemes 5 , Mátyás Pajkos 6 , Tamas Lazar 7,8 , Samuel Peña-Dı́az 9,10 , Jaime Santos 9,10 , Veronika Ács 4 , Nazanin Farahi 7,8 , Erzsébet Fichó 4,11 , Maria Cristina Aspromonte 12,13 , Claudio Bassot 14 , Anastasia Chasapi 15 , Norman E. Davey 16 , Radoslav Davidović 17 , Laszlo Dobson 3,4 , Arne Elofsson 14 , Gábor Erdős 6 , Pascale Gaudet 18 , Michelle Giglio 19 , Juliana Glavina 5 , Javier Iserte 20 , Valentı́n Iglesias 9,10 , Zsófia Kálmán 21 , Matteo Lambrughi 22 , Emanuela Leonardi 12,13 , Sonia Longhi 23 , Sandra Macedo-Ribeiro 24,25 , Emiliano Maiani 22 , Julia Marchetti 26 , Cristina Marino-Buslje 20 , Attila Mészáros 7,8 , Alexander Miguel Monzon 2 , Giovanni Minervini 2 , Suvarna Nadendla 19 , Juliet F. Nilsson 23 , Marian Novotný 27 , Christos A. Ouzounis 15,28 , Nicolás Palopoli 26 , Elena Papaleo 22,29 , Pedro José Barbosa Pereira 24,25 , Gabriele Pozzati 14 , Vasilis J Promponas 30 , Jordi Pujols 9,10 , Alma Carolina Sanchez Rocha 31 , Martin Salas 26 , Luciana Rodriguez Sawicki 26 , Eva Schad 4 , Aditi Shenoy 14 , Tamás Szaniszló 6 , Konstantinos D. Tsirigos 32 , Nevena Veljkovic 17 , Gustavo Parisi 26 , Salvador Ventura 9,10,33 , Zsuzsanna Dosztányi 6 , Peter Tompa 4,7,8 , Silvio C.E. Tosatto 2,* and Damiano Piovesan 2,* 1 Institute of Biomembranes, Bioenergetics and Molecular Biotechnologies, National Research Council (CNR-IBIOM), Bari, Italy, 2 Department of Biomedical Sciences, University of Padova, Padova, Italy, 3 Structural and Computational Biology Unit, European Molecular Biology Laboratory, Heidelberg 69117, Germany, 4 Institute of Enzymology, Research Centre for Natural Sciences, Budapest 1117, Hungary, 5 Instituto de Investigaciones Biotecnológicas (IIBiO-CONICET), Universidad Nacional de San Martı́n, Av. 25 de Mayo y Francia, CP1650 Buenos Aires, Argentina, 6 Department of Biochemistry, Eötvös Loránd University, Pázmány Péter stny 1/c, Budapest H-1117, Hungary, 7 VIB-VUB Center for Structural Biology, Vlaams Instituut voor Biotechnology, Brussels, Belgium, 8 Structural Biology Brussels (SBB), Bioengineering Sciences Department, Vrije Universiteit Brussel (VUB), Brussels, Belgium, 9 Institut de Biotecnologia i Biomedicina, Universitat Autònoma de Barcelona, Barcelona, Spain, 10 Departament de Bioquı́mica i Biologia Molecular, Universitat Autònoma de Barcelona, Barcelona, Spain, 11 Cytocast Kft., Vecsés, Hungary, 12 Department of Woman and Child Health, University of Padova, Padova, Italy, 13 Pediatric Research Institute, Città della Speranza, Padova, Italy, 14 Science for Life Laboratory, Department of Biochemistry and Biophysics, Stockholm University, 171 21 Solna, Sweden, 15 Biological Computation & Process Laboratory, Chemical Process & Energy Resources Institute, Centre for Research & Technology Hellas, Thermi, Thessalonica 57001, Greece, 16 Institute of Cancer Research, Chester Beatty Laboratories, 237 Fulham Rd, Chelsea, London, UK, 17 Laboratory for Bioinformatics and Computational Chemistry, Vinča Institute of Nuclear Sciences, National Institute of the Republic of Serbia, University of Belgrade, 11000 Belgrade, Serbia, 18 Swiss-Prot group, SIB Swiss Institute of Bioinformatics, Geneva, Switzerland, 19 Institute for Genome Sciences, University of Maryland School of Medicine 670 W. Baltimore St., Baltimore, MD 21201, USA, 20 Bioinformatics Unit, Fundación Instituto Leloir, Buenos Aires, * To whom correspondence should be addressed. Tel: +39 049 827 6269; Email: damiano.piovesan@unipd.it Correspondence may also be addressed to Silvio C.E. Tosatto. Email: silvio.tosatto@unipd.it  C The Author(s) 2021. Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Nucleic Acids Research, 2022, Vol. 50, Database issue D481 C1405BWE, Argentina, 21 Faculty of Information Technology and Bionics, Pázmány Péter Catholic University, Práter u. 50/A, 1083 Budapest, Hungary, 22 Cancer Structural Biology, Danish Cancer Society Research Center, Strandboulevarden 49, 2100 Copenhagen, Denmark, 23 Lab. Architecture et Fonction des Macromolécules Biologiques (AFMB), UMR 7257, Aix Marseille University and Centre National de la Recherche Scientifique (CNRS), 163 Avenue de Luminy, Case 932, 13288, Marseille, France, 24 Instituto de Biologia Molecular e Celular (IBMC), Universidade do Porto, 4200-135 Porto, Portugal, 25 Instituto de Investigação e Inovação em Saúde (i3S), Universidade do Porto, 4200-135 Porto, Portugal, 26 Departamento de Ciencia y Tecnologı́a, Universidad Nacional de Quilmes - CONICET, Bernal, Buenos Aires B1876BXD, Argentina, 27 Dep. of Cell Biology, Faculty of Science, Vinicna 7, 128 43, Prague, Czech Republic, 28 Biological Computation & Computational Biology Group, Artificial Intelligence & Information Analysis Lab, Department of Computer Science, Aristotle University of Thessalonica, Thessalonica 54124, Greece, 29 Cancer Systems Biology, Section for Bioinformatics, Department of Health and Technology, Technical University of Denmark, Lyngby, Denmark, 30 Bioinformatics Research Laboratory, Department of Biological Sciences, University of Cyprus, Nicosia, Cyprus, 31 Department of Cell Biology, Faculty of Science, Charles University, BIOCEV, Prague , Czech Republic, 32 European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Genome Campus, Hinxton, UK and 33 ICREA, Barcelona, Spain Received September 15, 2021; Revised October 15, 2021; Editorial Decision October 18, 2021; Accepted October 20, 2021 ABSTRACT As new disorder prediction methods keep emerging (5–7), CAID takes on monitoring the field in real time, aiming to The Database of Intrinsically Disordered Proteins establish dependable standards. This ambition has a special (DisProt, URL: https://disprot.org) is the major repos- caveat, as predicting and identifying regions in IDPs/IDRs itory of manually curated annotations of intrinsically that engage in functional interactions remains a significant disordered proteins and regions from the literature. challenge (6,8). We report here recent updates of DisProt version The prediction of such functional regions may support 9, including a restyled web interface, refactored In- and inspire dedicated experimental approaches. The Eu- trinsically Disordered Proteins Ontology (IDPO), im- karyotic Linear Motif (ELM) database is a primary reposi- provements in the curation process and significant tory of such data (9). Of similar ambition pursued along dif- content growth of around 30%. Higher quality and ferent lines, the database of fuzzy complexes, FuzDB, com- consistency of annotations is provided by a newly piles experimentally observed fuzzy protein complexes, in which intrinsic disorder is maintained upon partner inter- implemented reviewing process and training of cu- action, directly impacting biological function (10). An in- rators. The increased curation capacity is fostered structive example of this behavior is the extremely tight and by the integration of DisProt with APICURON, a ded- functional interaction between the disordered histone H1 icated resource for the proper attribution and recog- and its chaperone prothymosin-␣, which retain their highly nition of biocuration efforts. Better interoperability is dynamic, fully disordered state in that complex (11). Struc- provided through the adoption of the Minimum Infor- tural interpretation of such a behavior may be assisted by mation About Disorder (MIADE) standard, an active data in the Protein Ensemble Database (PED) that encom- collaboration with the Gene Ontology (GO) and Evi- passes experimentally determined structural ensembles of dence and Conclusion Ontology (ECO) consortia and IDPs/IDRs (12). the support of the ELIXIR infrastructure. There appears to be a consensus that these and other types of data and approaches will act in synergy to drive the field forward (13), toward reaching a better structural- INTRODUCTION functional understanding of the ‘disorderome’. Combined Whereas our traditional view of protein function is rooted with data in the MobiDB database, which provides predic- in the model of proteins assuming a stable structure, a well- tions and annotations for all IDPs/IDRs (14), this effort defined 3D fold, it is now >20 years since the concept of is also critical for integrating disorder-related information structural disorder of proteins has been proposed (1,2). The into other data resources, such as UniProtKB (15) or PDBe existence and functional importance of intrinsically disor- (16). dered proteins/regions (IDPs/IDRs) is now generally ac- Not surprisingly, the field is also getting strong impe- cepted (3), with >1500 PubMed publications mentioning tus from traditional structural biology: as missing regions disordered proteins every year. from solved structures constitute a good proxy for struc- The prediction of protein disorder from sequence, for ex- tural disorder, improvements in structure determination ample, has always been an area of continuous activity. Re- techniques boost the identification of structural disorder. cently, it has received a boost with the establishment of the Cryo-electron microscopy has now advanced to a state that Critical Assessment of Protein Intrinsic Disorder predic- yields structures at atomic resolution (17). Similarly, the tion (CAID) experiment as a community-wide blind test to recent success of computational structure prediction by compare state-of-the-art approaches to predict disorder (4). AlphaFold 2 (18) in overtaking the Critical Assessment D482 Nucleic Acids Research, 2022, Vol. 50, Database issue of Structure Prediction competition, CASP14 (19), cannot glutamate (polyE), poly-lysine (polyK) and poly-arginine be dismissed. AlphaFold’s 2-based structure predictions of (polyR) regions, have been added based on indirect evi- the entire human proteome (20) resulted in ‘only’ 58% of dence. These regions are always disordered but experimen- residues confidently covered with predicted structures, sug- tal evidence is provided for only a handful of cases or for gesting that structural disorder may be even more preva- engineered fragments. For example the polyE repeat in the lent in proteomes than previously thought (21). Whereas PEVK region of human titin has a Stokes radius 2–4 times it remains to be confirmed if certain AlphaFold 2 metrics larger than expected based on its molecular mass, and it can be harnessed for disorder prediction (20), it is without shows a minimum at 200 nm in its circular dichroism (CD) doubt that these achievements will help us navigate further spectrum (32). In order to be included in DisProt a repeated through the ‘dark proteome’ (22). region must be predicted by MobiDB-lite (33) (and be avail- Disordered proteins are often involved in disease, yet able in MobiDB) as a negative or positive polyelectrolyte they represent a largely unexplored target for drug devel- subregion (14,32) and be at least 10 residues long. In Dis- opment (23,24). Recent successes in the field, i.e. targeting Prot 9, such cases are highlighted by a specific evidence c-Myc (25), androgen receptor (26) or alpha-synuclein (27), code, curator inference from database (ECO:0007636). encourage us to adhere to this ambition. The concept of The distribution of regions based on the experimental de- liquid-liquid phase separation (LLPS) leading to the forma- tection method is shown in Figure 1. In accordance with tion of membraneless organelles (MLOs) raises hope that the recent improvements that recognize cryo-EM as a well- IDP/IDR function and therapeutic targeting can now be established technique in structural biology (17), DisProt in- approached from a novel angle (28,29). cludes a total of 149 IDRs annotated by this method. To reflect the steady progress of the protein disorder Figure 2 shows the length distribution of regions anno- field, it is important to update and upgrade DisProt as tated using a specific experimental technique. Atomic reso- one of the primary resources of manually curated, experi- lution techniques such as X-ray (ECO:0006187) and NMR mentally confirmed protein disorder. The previous release (ECO:0006252) dominate the experiments used to charac- of the database, DisProt 8 (30) contained about 1500 en- terize short (<100 residues) disordered regions, while other tries and 3500 disordered protein regions. In the current re- complementary methods and far-UV CD (ECO:0006179) lease, DisProt 9, not only did we increase these numbers, are the techniques mostly used in detecting and character- but also improved the reliability of entries by introducing izing longer IDRs. a reviewing process. In addition, a great effort was allo- Figure 3 illustrates the amino acid composition and cated to training activities by providing DisProt biocurators fold increase (enrichment) of DisProt proteins, as com- with detailed curation guidelines and virtual training ses- pared to TrEMBL distribution (release 2021 03). The fold sions and a published protocol describing how to explore increase, which is the difference between DisProt and manually curated annotations in DisProt (13), alongside TrEMBL frequencies normalized by TrEMBL, highlights the dissemination of new database content in a dedicated amino acids which are over- and under-represented in DisProt blog (https://disprot.org/blog/) and Twitter account DisProt IDRs. DisProt regions are enriched in disorder- (https://twitter.com/disprot db). DisProt 9 presents a new promoting residues (Q, K, P, E, S), mostly charged and hy- graphical interface and updated features, such as the in- drophilic, while the hydrophobic, order-promoting residues tegration of two ontologies and the connection with API- (F, A, V, I, L), as well as arginine (R) are depleted and aro- CURON, a database to credit and acknowledge the work matic residues (W, Y, F) are strongly depleted. This amino of biocurators (31). With these improvements, DisProt con- acid distribution is in line with previously published results tinues to be a primary resource of protein disorder for the (34,35). structural-molecular biology community. Thematic datasets PROGRESS AND NEW FEATURES Starting 2020, DisProt periodically releases ‘thematic Database content datasets’ describing biological areas where IDPs play a cru- DisProt 9 includes 2038 protein entries and 4,477 pieces of cial role. The first two thematic DisProt datasets illustrate evidence of state transitions, interactions and functions, fea- the role of intrinsic disorder in unicellular toxins and antitox- turing a 30% increase over the last release (DisProt 8), along ins (December 2020) and extracellular matrix proteins (June with 2578 publications annotated, accounting for a 28% in- 2021), by collecting carefully curated examples of IDPs and crease compared with DisProt 8. In addition, about 14% of IDRs involved in the toxin-antitoxin system and in the ex- the annotations, corresponding to 237 entries, have been re- tracellular matrix composition and function, respectively. viewed and validated by an expert biocurator, e.g. the High DisProt thematic datasets are easily accessible from the Dis- mobility group protein HMG-I/HMG-Y (DP00040) has Prot home page under the ‘Datasets’ section. All the entries been thoroughly reannotated by revising spurious annota- included in these datasets are tagged with the name of the tions and integrating new functions, such as DNA bending theme and it is possible to download them as single files. (GO:0008301, IDPO:00514), DNA binding (GO:0003677, IDPO:00065), RNA binding (GO:0003723, IDPO:00066) Ontology and protein binding (GO:0005515, IDPO:00063). In DisProt 9, annotations of amino acid repeats dis- In DisProt 9, the previous Disordered Ontology (DO) has playing the typical properties of IDPs, specifically poly- been renamed Intrinsically Disordered Protein Ontology Nucleic Acids Research, 2022, Vol. 50, Database issue D483 Figure 1. Number of experimental evidence of disorder used in DisProt by year of publication. Different colors correspond to different experimental techniques as reported in the corresponding publications. Publications older than 1990 are grouped in the first bar. Figure 2. Region length distribution of pieces of evidence inferred using various experimental techniques in DisProt. Data in the main plot starts at length 40, the full distribution is shown in the inset plot. (IDPO) and has been completely refactored. The ‘experi- GO, thanks to an active collaboration between the DisProt mental method’ branch has been mapped one-to-one to a and GO consortia. subset of the Evidence and Conclusion Ontology (ECO) All active functions, which exert their effect via molecular (36), which was recently updated with new disorder-specific recognition on another interactor, are now annotated using terms, e.g. IDPO:00125 maps to the corresponding small- GO terms, while self-regulatory functions (self-regulation angle X-ray scattering (ECO:0006182) ECO term. In addi- or self-assembly via protein interactions in cis) and func- tion, several new non-IDP-specific terms were introduced to tions directly arising from the disordered state (entropic cover a plethora of missing techniques, e.g. bait-prey pro- chains) are annotated by specific IDPO terms since cor- tein pull-down evidence (ECO:0006249) is widely used to responding terms are not yet available in GO. Entropic assess interactions and functions of IDPs and IDRs. bristle, entropic spring and entropic clock terms have now Similarly, all interaction terms and some function terms been merged under the entropic chain term. The Flexible N- have been mapped to the Gene Ontology (GO) (37–39). Di- terminal tail and flexible C-terminal tail terms were intro- verse modes of action of IDPs have been completely revised duced to highlight the presence of disorder in protein ter- to provide the highest possible level of detail in the char- minal regions. acterization of disorder functions. Despite GO being de- The DisProt 9 website incorporates a new dedicated page signed to annotate the whole protein, several of its terms are for each IDPO term, e.g. https://disprot.org/idpo/IDPO: well suited to describe IDR behaviour, in particular those in 00501 for the entropic chain term. IDPO term pages include the molecular function branch. IDPO terms were merged, the identifier, name and definition of the term, its relation- mapped to existing GO terms or directly integrated into ships with other terms and, when available, cross-references D484 Nucleic Acids Research, 2022, Vol. 50, Database issue Figure 3. Amino acid composition of DisProt regions. Amino acids are sorted by the Kyte-Doolittle hydrophobicity scale. The amino acid frequency is calculated considering only disordered residues. The enrichment is calculated and normalized over the TrEMBL database frequencies (release 2021 03). Color intensity is proportional to bar height. to external ontologies, e.g. Gene Ontology. Moreover, they Biocuration and APICURON list all entries annotated with that term. DisProt annotations are provided by both professional and The adoption of stable and well established ontologies, community biocurators. The whole community of biocu- GO and ECO, plays a crucial role by allowing the curators rators is supported by a team of senior biocurators who to effortlessly expand the coverage of functions and exper- check and validate annotations and deliver training ma- imental techniques available in DisProt. Since its partner- terial including face-to-face activities and a detailed cu- ship with the GO and ECO consortia, DisProt has become ration manual. In order to properly attribute the cura- active in the definition of new terms that fulfill ontology- tion effort, DisProt is now connected to APICURON (31). specific rules and constraints. Requests for new terms com- APICURON collects and tracks biocuration events from ing from the IDP community are welcome but must un- manually curated resources and implements gamification dergo GO and ECO approval. In the long run, all terms concepts, i.e. badges, medals and leaderboards, to pro- used by DisProt will be mapped to well established ontolo- mote biocurator engagement. The DisProt website inte- gies. grates an APICURON widget (https://disprot.org/release- notes) which provides the ranking, number of activities and Experimental ambiguity and MIADE scores of the curators. In APICURON, all DisProt biocu- rators have a dedicated profile page that groups all the A number of DisProt annotations come from experiments achieved badges and medals, along with their curation ac- performed under extreme conditions. These annotations, tivity. which were tagged as ambiguous in the previous DisProt Additional annotations can be provided directly by the version, are now described using the Minimum Informa- DisProt users through a ‘Contact us’ page. In DisProt 9, tion About Disorder Experiments (MIADE) guidelines along with the ‘Leave a comment’ tab for feedback on the (https://www.psidev.info/groups/intrinsically-disordered- site experience, a new ‘Submit a new annotation’ section has proteins) (40). MIADE defines the minimal fundamental been added. parameters that unambiguously characterize a disorder- related experiment. With MIADE, it is possible to evaluate Implementation and compare experimental evidence coming from other resources adopting the same standard objectively. DisProt 9 adopts the Document Versioning Pattern to pre- In DisProt 9, MIADE is implemented by including the cisely track all changes between different versions of the following annotation fields: (i) sequence construct features, same record. This improves reproducibility, simplifies track- e.g. the exact sequence from the experiment along with its ing of curation activities and allows the user to compare dif- modifications, such as PTMs and mutations, (ii) the experi- ferent versions of an annotated entry directly on the web mental conditions, i.e. pH, temperature, pressure and redox site. Each protein has a history page and each IDR has an potential and (iii) the experimental components, e.g. small- assigned version number. Whenever a curator modifies an molecules and membranes, along with values, deviation and existing region, the version counter increases by one. additional variables. Further specifications are provided in DisProt exposes disordered region descriptions with the help page on the DisProt website. Bioschemas metadata, including disordered region Nucleic Acids Research, 2022, Vol. 50, Database issue D485 boundaries and type of disorder. The information content inition of MIADE guidelines, creation of a standard con- from the entry pages is aggregated together with comple- trolled vocabulary to define conformational states (in col- mentary resources (MobiDB and PED) to the IDPcentral laboration with the structural biology community). From a registry and knowledgebase (41). technological point of view, DisProt is working to meet the The DisProt web application source code underwent a format standards to export its annotations to core data re- major upgrade, updating Angular version 12 and transi- sources such as GO, PDB, IntAct and UniProt. The long- tioning to Bootstrap version 5. The DisProt interface is term maintenance of DisProt is guaranteed by its central now more intuitive, thanks to a minimalistic, consistent de- role within the European Union’s Horizon 2020 IDPfun sign between different components. The upgraded curation program and the ELIXIR IDP Community, the reference interface allows for more efficient work from biocurators, scientific communities involved in the study of intrinsically while raising the standards of curated data quality thanks disordered proteins. to real-time syntax checks and cross-validation with vari- ous third-party data resources. DATA AVAILABILITY The data that support the findings of this study are openly Outreach activities available in DisProt at https://disprot.org/. DisProt has an active social media presence on Twit- ter (https://twitter.com/disprot db). This account promotes ACKNOWLEDGEMENTS communication with users and experts in the field. It pro- vides updated information about the latest release, webi- DisProt is a service of the Italian node of ELIXIR. This nars and training activities. Finally, a dedicated blog web- project has received funding from the European Union’s site (https://disprot.org/blog) contains descriptive extended Horizon 2020 research and innovation programme under posts about thematic datasets, general statistics and novel, grant agreement no. 778247 and no. 952334. This project interesting annotations. Training material for the users is has received funding from the Italian Ministry of University provided via a published protocol describing how to explore and Research (MIUR), PRIN 2017 under grant agreement manually curated annotations in DisProt (13). Training ma- no. 2017483NH8. terial for the curators is offered via recorded webinars and an updated user manual. FUNDING Italian Ministry of University and Research (MIUR) CONCLUSIONS AND FUTURE WORK [2017483NH8]; European Union’s Horizon 2020 [778247 to DisProt is the gold standard for IDP/IDR annotations, and S.C.E.T.]; European Union’s Horizon 2020 research and in- serves the community as a fundamental resource that drives novation programme under the Marie Skłodowska-Curie biological hypotheses, experimental design and the train- [842490 MIMIC) to B.M.]; Tempus Public Foundation ing and benchmarking of disorder and function predictors. [158534]; NRDI Office [FK128133] of the Hungarian gov- Compared to the previous version, it has improved data ac- ernment (to R.P.); National Agency for the Promotion of cessibility and quality, and significantly increased annota- Science and Technology (ANPCyT) [PICT-2017-1924 to tion volume. The content is updated frequently and is now L.B.C.]; Spanish Ministry of Science and Innovation via a more focused on function. A team of expert reviewers vali- doctoral grant [FPU17/01157 to J.S.]; European Union’s date annotations provided by the community and continu- Horizon 2020 research and innovation programme un- ously check the literature for novel experimental evidence. der the Marie Sklodowska-Curie [101028908 (SMILE) DisProt is well connected to other databases and con- to L.D.]; Swedish Research Council for Natural Science sortia, and is active in the development of new standards [VR-2016-06301]; Swedish E-science Research Center (to and ontologies. The Intrinsically Disordered Proteins Ontol- A.E.); National Human Genome Research Institute grant ogy (IDPO) has been refactored and systematically cross- [U41 HG02273 to P.G.]; CNRS (to S.L.); Infectiopôle referenced with Gene Ontology (GO) and Evidence and Con- Sud post-doctoral fellowship (to J.F.N.); ELIXIR CZ Re- clusion Ontology (ECO). Experimental setup is now cap- search Infrastructure [ID LM2018131, MEYS CR]; Min- tured in an unambiguous and structured way by implement- istry of Education, Science, and Technological Develop- ing the Minimum Information About Disorder (MIADE) ment of the Republic of Serbia (to R.D., N.V.); Univer- standard. sidad Nacional de Quilmes [PUNQ-2019-1309/19]; Na- The DisProt technological infrastructure has been re- tional Agency for the Promotion of Science and Tech- newed to improve reproducibility by implementing exhaus- nology (ANPCyT) [PICT-2018-3457 to G.P.]; Hungarian tive versioning of all entries. The community of curators Scientific Research Fund (OTKA) [K124670, K131702]; is engaged, structured, well trained and continuously up- VUB Spearhead grant [SRP51, 2019–24 to P.T.]; Elixir- dated. Finally, DisProt is now connected with APICURON GR [MIS 5002780 to A.C., C.A.O.]; J.M. is a PhD stu- to provide live tracking and proper attribution of the cura- dent; J.G. is a postdoc and GP, C.M.-B., J.I., L.B.C. are tion effort. researchers of the National Research Council (CONICET) DisProt is committed to the reduction of false negative of Argentina; Cancer Research UK Senior Cancer Re- annotations, resolution of inconsistent annotations (e.g. be- search Fellowship [C68484/A28159 to N.E.D.]; University tween close homologs) and growth of functional descrip- of Cyprus [Computational approaches towards mechanistic tions. At the consortium level, DisProt is active on the def- insights and improved detection of functional LIR-motifs D486 Nucleic Acids Research, 2022, Vol. 50, Database issue in selective autophagy receptors and adaptors. (idLIR)] in- 18. Jumper,J., Evans,R., Pritzel,A., Green,T., Figurnov,M., ternal grant (to V.J.P.); Hungarian Scientific Research Fund Ronneberger,O., Tunyasuvunakool,K., Bates,R., Žı́dek,A., (OTKA) [K129164, K139284 to Z.D.]; European Union’s Potapenko,A. et al. (2021) Highly accurate protein structure prediction with AlphaFold. Nature, 596, 583–589. Horizon 2020 research and innovation programme [952334 19. Pereira,J., Simpkin,A.J., Hartmann,M.D., Rigden,D.J., Keegan,R.M. PhasAGE to S.M.-R.]. Funding for open access charge: Eu- and Lupas,A.N. (2021) High-accuracy protein structure prediction in ropean Union’s Horizon 2020 research and innovation pro- CASP14. Proteins Struct. Funct. Bioinforma., gramme [778247]. https://doi.org/10.1002/prot.26171. 20. Tunyasuvunakool,K., Adler,J., Wu,Z., Green,T., Zielinski,M., Conflict of interest statement. None declared. Žı́dek,A., Bridgland,A., Cowie,A., Meyer,C., Laydon,A. et al. (2021) Highly accurate protein structure prediction for the human proteome. Nature, 596, 590–596. REFERENCES 21. Pancsa,R. and Tompa,P. (2012) Structural disorder in eukaryotes. 1. Romero,P., Obradovic,Z., Kissinger,C.R., Villafranca,J.E., PLoS One, 7, e34687. Garner,E., Guilliot,S. and Dunker,A.K. (1998) Thousands of 22. Bhowmick,A., Brookes,D.H., Yost,S.R., Dyson,H.J., proteins likely to have long disordered regions. Pac. Symp. Biocomput. Forman-Kay,J.D., Gunter,D., Head-Gordon,M., Hura,G.L., Pac. Symp. Biocomput., 1998, 437–448. Pande,V.S., Wemmer,D.E. et al. (2016) Finding our way in the dark 2. Wright,P.E. and Dyson,H.J. (2015) Intrinsically disordered proteins in proteome. J. Am. Chem. Soc., 138, 9730–9742. cellular signalling and regulation. Nat. Rev. Mol. Cell Biol., 16, 18–29. 23. Ruan,H., Sun,Q., Zhang,W., Liu,Y. and Lai,L. (2019) Targeting 3. van der Lee,R., Buljan,M., Lang,B., Weatheritt,R.J., intrinsically disordered proteins at the edge of chaos. Drug Discov. Daughdrill,G.W., Dunker,A.K., Fuxreiter,M., Gough,J., Gsponer,J., Today, 24, 217–227. Jones,D.T. et al. (2014) Classification of intrinsically disordered 24. Santofimia-Castaño,P., Rizzuti,B., Xia,Y., Abian,O., Peng,L., regions and proteins. Chem. Rev., 114, 6589–6631. Velázquez-Campoy,A., Neira,J.L. and Iovanna,J. (2020) Targeting 4. Necci,M., Piovesan,D., Predictors,C., Curators,D.P. and intrinsically disordered proteins involved in cancer. Cell. Mol. Life Tosatto,S.C.E. (2021) Critical assessment of protein intrinsic disorder Sci. CMLS, 77, 1695–1707. prediction. Nat. Methods, 18, 472–481. 25. Boike,L., Cioffi,A.G., Majewski,F.C., Co,J., Henning,N.J., 5. Hu,G., Katuwawala,A., Wang,K., Wu,Z., Ghadermarzi,S., Gao,J. Jones,M.D., Liu,G., McKenna,J.M., Tallarico,J.A., Schirle,M. et al. and Kurgan,L. (2021) flDPnn: accurate intrinsic disorder prediction (2021) Discovery of a functional covalent ligand targeting an with putative propensities of disorder functions. Nat. Commun., 12, intrinsically disordered cysteine within MYC. Cell Chem. Biol., 28, 4438. 4–13. 6. Hanson,J., Paliwal,K.K., Litfin,T. and Zhou,Y. (2019) 26. Ban,F., Leblanc,E., Cavga,A.D., Huang,C.-C.F., Flory,M.R., SPOT-Disorder2: improved protein intrinsic disorder prediction by Zhang,F., Chang,M.E.K., Morin,H., Lallous,N., Singh,K. et al. ensembled deep learning. Genomics Proteomics Bioinformatics, 17, (2021) Development of an androgen receptor inhibitor targeting the 645–656. N-Terminal domain of androgen receptor for treatment of castration 7. Oberti,M. and Vaisman,I.I. (2020) cnnAlpha: protein disordered resistant prostate cancer. Cancers, 13, 3488. regions prediction by reduced amino acid alphabets and 27. Tóth,G., Neumann,T., Berthet,A., Masliah,E., Spencer,B., Tao,J., convolutional neural networks. Proteins, 88, 1472–1481. Jobling,M.F., Gardai,S.J., Bertoncini,C.W., Cremades,N. et al. (2019) 8. Mészáros,B., Erdős,G. and Dosztányi,Z. (2018) IUPred2A: Novel small molecules targeting the intrinsically disordered structural context-dependent prediction of protein disorder as a function of ensemble of ␣-synuclein protect against diverse ␣-synuclein mediated redox state and protein binding. Nucleic. Acids. Res., 46, dysfunctions. Sci. Rep., 9, 16947. W329–W337. 28. Mathieu,C., Pappu,R.V. and Taylor,J.P. (2020) Beyond aggregation: 9. Kumar,M., Gouw,M., Michael,S., Sámano-Sánchez,H., Pancsa,R., pathological phase transitions in neurodegenerative disease. Science, Glavina,J., Diakogianni,A., Valverde,J.A., Bukirova,D., Čalyševa,J. 370, 56–60. et al. (2020) ELM––the eukaryotic linear motif resource in 2020. 29. Cai,D., Liu,Z. and Lippincott-Schwartz,J. (2021) Biomolecular Nucleic. Acids. Res., 48, D296–D306. condensates and their links to cancer progression. Trends Biochem. 10. Miskei,M., Antal,C. and Fuxreiter,M. (2017) FuzDB: database of Sci., 46, 535–549. fuzzy complexes, a tool to develop stochastic structure-function 30. Hatos,A., Hajdu-Soltész,B., Monzon,A.M., Palopoli,N., Álvarez,L., relationships for protein complexes and higher-order assemblies. Aykac-Fas,B., Bassot,C., Benı́tez,G.I., Bevilacqua,M., Chasapi,A. Nucleic Acids Res., 45, D228–D235. et al. (2020) DisProt: intrinsic protein disorder annotation in 2020. 11. Borgia,A., Borgia,M.B., Bugge,K., Kissling,V.M., Heidarsson,P.O., Nucleic Acids Res., 48, D269–D276. Fernandes,C.B., Sottini,A., Soranno,A., Buholzer,K.J., Nettels,D. 31. Hatos,A., Quaglia,F., Piovesan,D. and Tosatto,S.C.E. (2021) et al. (2018) Extreme disorder in an ultrahigh-affinity protein APICURON: a database to credit and acknowledge the work of complex. Nature, 555, 61–66. biocurators. Database J. Biol. Databases Curation, 2021, baab019. 12. Lazar,T., Martı́nez-Pérez,E., Quaglia,F., Hatos,A., Chemes,L.B., 32. Duan,Y., DeKeyser,J.G., Damodaran,S. and Greaser,M.L. (2006) Iserte,J.A., Méndez,N.A., Garrone,N.A., Saldaño,T.E., Marchetti,J. Studies on titin PEVK peptides and their interaction. Arch. Biochem. et al. (2021) PED in 2021: a major update of the protein ensemble Biophys., 454, 16–25. database for intrinsically disordered proteins. Nucleic Acids Res., 49, 33. Necci,M., Piovesan,D., Clementel,D., Dosztányi,Z. and D404–D411. Tosatto,S.C.E. (2020) MobiDB-lite 3.0: fast consensus annotation of 13. Quaglia,F., Hatos,A., Piovesan,D. and Tosatto,S.C.E. (2020) intrinsic disorder flavors in proteins. Bioinformatics, 36, 5533–5534. Exploring manually curated annotations of intrinsically disordered 34. Dunker,A.K., Lawson,J.D., Brown,C.J., Williams,R.M., Romero,P., proteins with disprot. Curr. Protoc. Bioinforma., 72, e107. Oh,J.S., Oldfield,C.J., Campen,A.M., Ratliff,C.M., Hipps,K.W. et al. 14. Piovesan,D., Necci,M., Escobedo,N., Monzon,A.M., Hatos,A., (2001) Intrinsically disordered protein. J. Mol. Graph. Model., 19, Mičetić,I., Quaglia,F., Paladin,L., Ramasamy,P., Dosztányi,Z. et al. 26–59. (2021) MobiDB: intrinsically disordered proteins in 2021. Nucleic 35. Necci,M., Piovesan,D. and Tosatto,S.C.E. (2018) Where differences Acids Res., 49, D361–D367. resemble: sequence-feature analysis in curated databases of 15. UniProt Consortium (2019) UniProt: a worldwide hub of protein intrinsically disordered proteins. Database J. Biol. Databases knowledge. Nucleic Acids Res., 47, D506–D515. Curation, 2018, bay127. 16. PDBe-KB consortium (2020) PDBe-KB: a community-driven 36. Giglio,M., Tauber,R., Nadendla,S., Munro,J., Olley,D., Ball,S., resource for structural and functional annotations. Nucleic Acids Mitraka,E., Schriml,L.M., Gaudet,P., Hobbs,E.T. et al. (2019) ECO, Res., 48, D344–D353. the Evidence & Conclusion Ontology: community standard for 17. Nakane,T., Kotecha,A., Sente,A., McMullan,G., Masiulis,S., evidence information. Nucleic Acids Res., 47, D1186–D1194. Brown,P.M.G.E., Grigoras,I.T., Malinauskaite,L., Malinauskas,T., 37. Gene Ontology Consortium (2021) The Gene Ontology resource: Miehling,J. et al. (2020) Single-particle cryo-EM at atomic resolution. enriching a GOld mine. Nucleic Acids Res., 49, D325–D334. Nature, 587, 152–156. Nucleic Acids Research, 2022, Vol. 50, Database issue D487 38. Ashburner,M., Ball,C.A., Blake,J.A., Botstein,D., Butler,H., 40. Davey,N.E., Babu,M.M., Blackledge,M., Bridge,A., Cherry,J.M., Davis,A.P., Dolinski,K., Dwight,S.S., Eppig,J.T. et al. Capella-Gutierrez,S., Dosztanyi,Z., Drysdale,R., Edwards,R.J., (2000) Gene ontology: tool for the unification of biology. The Gene Elofsson,A., Felli,I.C. et al. (2019) An intrinsically disordered Ontology Consortium. Nat. Genet., 25, 25–29. proteins community for ELIXIR. F1000Research, 8, 1753. 39. Binns,D., Dimmer,E., Huntley,R., Barrell,D., O’Donovan,C. and 41. Gray,A.J.G., Papadopoulos,P., Mičetić,I. and Hatos,A. (2021) Apweiler,R. (2009) QuickGO: a web-based tool for Gene Ontology Exploiting bioschemas markup to populate IDPcentral. searching. Bioinforma. Oxf. Engl., 25, 3045–3046. BioHackrXiv, https://doi.org/10.37044/osf.io/v3jct.