Nucleic Acids Research, 2025, gkaf1258

[Link]
Database issue

RBP2GO 2.0: Integrating disease associations and

Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

sequence features to explore RNA-binding protein
functions
Simona Cantarella1 ,† , Jule Neffe1 ,† , Malte Hermes1,2 , Fabio Rauscher1 , Robert Schwarz1 ,
Praarthanaa Jaisankar1 , Enno Schäfer1 , Maiwen Caudron-Herger 1 ,*
1
Research group “RNA-Protein Complexes & Cell Proliferation”, German Cancer Research Center (DKFZ), Heidelberg 69120, Germany
2
Present address: Department of General, Visceral and Transplantation Surgery, University of Heidelberg, Heidelberg, Germany
∗
To whom correspondence should be addressed. Tel: +49 6221 424383; Fax: +49 6221 424384; Email: [Link]@[Link]
†
These authors contributed equally to this work

Abstract
RNA-binding proteins (RBPs) play key roles in a wide range of biological processes and human diseases. Here, we updated RBP2GO, a com-
prehensive resource on RBPs, their binding partners, and functions across 13 species. The database RBP2GO 2.0 now provides additional
information on disease ontology to better relate RBP functions to their impact in human diseases. It also integrates knowledge from new
proteome-wide RBP studies and provides specific information on sequence features through intuitive cartoon-style representations, such as
associated RNA-binding peptides, when available. Existing eCLIP/iCLIP datasets from ENCODE have been linked to the corresponding proteins,
along with RNA dependence information from the R-DeeP database. Protein characterization is further supported by two scores, the RBP2GO
Score and the RBP2GO Composite Score, which reflect the probability that the protein binds to RNA. The scores are now visualized using
violin plot distributions for the whole proteome of each species. The redesigned user interface enables intuitive searches for protein names,
gene ontology terms, disease ontology terms, and protein domains directly from the Home page and provides extended, versatile search op-
tions via the Advanced Search module. RBP2GO 2.0 serves as a valuable tool for investigating new RBP functions and is publicly accessible at
[Link]

Graphical abstract

Introduction RBPs include protein ubiquitination [6] in the context of im-

RNA-binding proteins (RBPs) are essential regulators of di- mune response, protein multimerization in the context of au-
verse biological processes. Traditionally, RBPs are best known tophagy [7], and riboregulation of unconventional RBPs—an
for their canonical functions in RNA processing, includ- emerging concept describing how RNA actively modulates
ing RNA splicing, nuclear-to-cytoplasmic transport, transla- the interactions, localization, and function of the interacting
tion regulation, and RNA decay [1−5]. Beyond these well- RBPs [7–11]. Moreover, accumulating evidence points to a
established roles, unconventional functions and properties of role of RBPs in the development of human diseases, including

Received: September 14, 2025. Revised: October 23, 2025. Accepted: October 23, 2025
© The Author(s) 2025. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License
([Link] which permits non-commercial re-use, distribution, and reproduction in any medium, provided the
original work is properly cited. For commercial re-use, please contact reprints@[Link] for reprints and translation rights for reprints. All other
permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact
[Link]@[Link].
 Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

Figure 1. RBP2GO 2.0 Home page. Screenshot of the design of RBP2GO 2.0 Home page, available at [Link]
Cantarella et al.
2
 Integrating disease associations and sequence features 3

cancer, diabetes, cardiovascular disorders, and neurodegener- Table 1. Statistics on RBP2GO 2.0 upgrade. The numbers show changes
ative diseases [12–16]. The wide range of biological functions in the distinct entries throughout the present and former versions of
RBP2GO. Bold numbers indicate changes compared to the previous ver-
covered by RBPs is accompanied by remarkable variability
sions of RBP2GO
in their modes of interaction with RNA. RBPs can interact
with RNA through structurally well-defined RNA-binding do- Type of information RBP2GO RBP2GO-2-Beta RBP2GO 2.0
mains (RBDs) [17, 18], such as the RNA recognition motif

Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

Species 13 13 13
(RRM) [19], the K homology (KH) domain [20], or the DEAD
Studies 53 53 72
box helicase domain [21]. Additionally, RNA-protein inter- Datasets 105 105 135
actions can be mediated by intrinsically disordered regions Proteins 176 940 176 896 176 896
(IDRs) [22] or can even occur in the absence of both RBDs RBPs 22 552 22 552 29 215
and IDRs [23]. Despite their biological importance, publicly RBDs 0 977 (+15 new) 992
available resources for RBPs remain limited and often targeted RBPs aspirants 0 7738 10 420
RNA-binding peptides 0 0 18 815
toward specific aspects such as eukaryotic RBPs [13, 24] or
Link to eCLIP/iCLIP 0 0 297
RNA-binding motifs [25]. Gene Ontology terms 27 871 27 871 27 871
The comprehensive integration of a large number of Disease Ontology terms 0 0 9319
proteome-wide studies, the possibility to conduct both
biology- and disease-oriented searches, and the inclusion of
information on RNA-binding peptides would greatly en- Table 2. Statistics on RBP2GO 2.0 RBPs. The values reflect changes in
hance the exploration of novel RBP functions. Therefore, we the number of datasets and RBPs across the 13 species from RBP2GO to
have developed RBP2GO 2.0 (Fig. 1), an updated version RBP2GO 2.0
of RBP2GO [26] and its online development RBP2GO-2-
RBP2GO RBP2GO
Beta [23]. We have incorporated data from newly published
RBP2GO RBP2GO 2.0 2.0
proteome-wide studies and integrated the disease ontology, Species (# datasets) (# RBPs) (# datasets) (# RBPs)
along with several new database features that expand its util-
ity. RBP2GO 2.0 hosts a total of 176 896 proteins, includ- Homo sapiens 43 6099 62 8757
ing 29 215 proteins that have been detected at least once Mus musculus 13 2897 18 4370
Saccharomyces 11 2155 12 2315
as RBP in high-throughput screens. Altogether, the RBP2GO
cerevisiae
2.0 database features an intuitive user interface that enables Drosophila 9 2979 9 2979
effective visualization of protein information, supported by melanogaster
links to external resources such as UniProt for basic pro- Arabidopsis 7 2352 7 2352
tein annotation [27], Gene Ontology (GO) for biological pro- thaliana
cesses, molecular functions, and cellular compartments [28], Caenorhabditis 6 1861 6 1861
elegans
and the STRING database [29] for protein interaction part-
Escherichia coli 3 1216 5 1323
ners. The connection to Disease Ontology (DO) terms via Plasmodium 4 1020 5 1619
the DISEASES 2.0 [30] and the DO-KB databases [31] fur- falciparum
ther enhances its utility by allowing users to identify pro- Danio rerio 3 311 3 311
teins implicated in one or more human diseases. A distinc- Salmonella 2 226 3 467
tive feature of RBP2GO remains the assignment of up to two Typhimurium
scores to each protein, reflecting its ability to bind to RNA: Leishmania 1 1162 2 2613
mexicana
(i) the RBP2GO Score is computed based on the frequency Trypanosoma 2 169 2 169
with which a protein and its binding partners are experimen- brucei
tally detected as RNA binding [26] and (ii) the RBP2GO Com- Leishmania 1 79 1 79
posite Score integrates the RBP2GO Score with available in- donovani
formation on RBDs or RNA-related InterPro family IDs [23,
32]. Importantly, the use of a new powerful Advanced Search
module across species—with flexible batch search options— several new publications and datasets since the original release
in combination with the RBP2GO Scores enables the identifi- of RBP2GO [26]. Manual curation of the underlying datasets
cation of candidate RBPs that have not yet been experimen- resulted in a total of 29 215 proteins listed as RBPs (Table 1).
tally detected or further validated. Lastly, RBP2GO 2.0 pro- The updated RBP datasets were compiled following the
vides easy access to available RNA-binding peptide data, fur- methodology described in the original RBP2GO publication
ther supporting the discovery of non-canonical RNA-binding [26]. Only proteome-wide studies, including both experimen-
sites. tal and in-silico approaches, were considered. The RBP in-
We anticipate that RBP2GO 2.0 will serve as a valuable clusion threshold was set to one, meaning that a protein ap-
resource for integrating the extensive information scattered pearing at least once in any dataset is classified as an RBP in
across multiple studies and will facilitate the discovery of RBP2GO. Detailed information about the methodology used
new RBPs and new RNA-protein functions in health and to identify each RBP is available from the “References &
disease contexts. RBP2GO 2.0 is freely accessible at https: Data” top menu of the database, where all datasets and studies
//[Link] without registration. are listed along with additional information.
Following the inclusion of 19 new studies in RBP2GO 2.0,
the numbers of datasets and RBPs associated with each species
Upgrades were updated accordingly (Table 2), as well as the frequency
Recent technical advances have enabled multiple high- of RBP detection in high-throughput screens, the RBP2GO
throughput approaches to efficiently identify RBPs, leading to Score, and the RBP2GO Composite Score. It is important to
4 Cantarella et al.

Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

Figure 2. RBP2GO 2.0 architecture. Four search boxes are available per quick search directly from the RBP2GO 2.0 Home page. Searching by a protein
name, a Gene Ontology (GO) term, or a Disease Ontology (DO) term returns a list of associated proteins (green). Selecting a protein from the list opens
the specific protein page for this protein. Searches initiated from the “Search InterPro Features” box return a list of related InterPro IDs with their related
description (orange). Selecting an InterPro ID from the list opens the Advanced Search with the InterPro ID already included in the corresponding search
field. The user can start the search or add additional filters prior to starting the search. Selecting a protein in the result list again opens the specific
protein page, offering various information categories about this protein (dark blue panel). New features in RBP2GO 2.0 include the DO search and
information with DO score, sequence features information, and links to external resources specific to human (Only Homo sapiens). Search inputs can be
provided as identifiers matching the UniProt, GO, DO, or InterPro databases, or as a keyword (not for DO). The terms HNRNPU, Mitosis, cancer, and
kinase were provided as an example in the figure.

note that a number of RBPs still require individual validation higher likelihood to bind to RNA. A definition of RBP aspi-
through targeted experiments—such as crosslinking immuno- rants and guidance on how to identify them is also available
precipitation (e.g. iCLIP or eCLIP) [33, 34]—to confirm their on the “About RBP2GO” page.
RNA-binding ability. RBP2GO 2.0 also contains RNA-binding peptide infor-
RBP2GO 2.0 was further expanded by integrating the mation when available for the protein of interest, compris-
new RBDs identified within the analysis accompanying the ing a total of 18 815 peptides manually curated from previ-
RBP2GO-2-Beta version [23], summing up to 992 RBDs. This ous studies [35–40] as well as links to 297 ENCODE eCLIP
enabled the identification of additional potential RBPs, re- and iCLIP datasets of interacting RNA transcripts [3, 34].
ferred to as RBP aspirants (now 10 420 in total), previously Finally, RBP2GO 2.0 integrates disease information through
defined as proteins containing an RBD but not yet experimen- 9319 disease ontology terms [30, 31] to efficiently support
tally detected as RBP (i.e. non-RBPs with RBDs) [23]. In the analyses of the proteins and RBPs in relation to human
Advanced Search tool, users can now identify RBP aspirants diseases.
by selecting “Only Non-Listed” proteins under the “Select The section “About RBP2GO” has been re-organized and
RBP” box and checking the “Only RBD-containing proteins” provides users with a Frequently Asked Questions (FAQs) sec-
option. Sorting the results by the RBP2GO Composite Score tion, aiming at facilitating the navigation and the efficient ex-
further facilitates the identification of RBP aspirants with a ploitation of the RBP2GO resources.
 Integrating disease associations and sequence features 5

Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

Figure 3. Disease Ontology search. The Advanced Search module allows users to explore the Disease Ontology database efficiently. Proteins associated
with a specific disease can be filtered to include only those identified as RNA-binding or RNA-dependent (indicated by the “Only RBP/R-DeeP”
checkbox, highlighted with the top black box in the left panel). Additionally, users can apply further criteria, such as an RBP2GO Composite Score greater
than 10 (bottom black box in the left panel). A comprehensive Disease Ontology tree is accessible via the folder tree icon in the left panel (see point 1),
enabling users to browse and add terms directly to their search (see points 2 and 3). For example, filtering for RNA-binding or RNA-dependent proteins
involved in cancer with an RBP2GO Composite Score ≥10 yields a list of 2247 proteins (point 4).

New features sources such as UniProt [27], InterPro [32], MobiDB [41] (dis-
ordered domains), Gene Ontology [28], STRING [29] (inter-
New user interface
acting proteins), UniRef [42] (protein homologs), R-DeeP [43,
The RBP2GO 2.0 user interface has been extensively simpli- 44] (RNA-dependent proteins), Disease Ontology [30, 31],
fied and re-designed since the original publication of RBP2GO and ENCODE [45] eCLIP/iCLIP datasets[3, 34]. The last three
[26] and its subsequent online development (RBP2GO-2-Beta) menus of the protein sidebar (R-DeeP, Disease Ontology, and
[23]. Searches for individual proteins, GO or DO terms, pro- ENCODE eCLIP/iCLIP) are specific to humans and disabled
tein domains, or protein families (as defined by the InterPro for the other species. The navigation through the protein re-
database [32]) can now be initiated directly from the Home sults, as well as to the protein page and back to the results, is
page as “Quick start.” The green-colored boxes return a list intuitive via dedicated buttons.
of one or more proteins that match the query protein name, Finally, the Advanced Search tool has been largely simpli-
GO or DO term, while the orange-colored box returns a list fied by integrating all species. It supports multi-term queries
of one or more InterPro IDs. When using the InterPro-related (batch search) and filtering of the results using various crite-
search option, each ID from the result list can be selected and ria. The Advanced Search module now appears as a sidebar
forwarded to the “Advanced Search” to either directly start accessible from both the Home and the “Search RBP2GO”
a search for proteins related to this ID or combine the ID top menus.
with further search parameters. Via the Advanced Search, a
list of proteins is returned (Fig. 2). A significant change as
compared to the previous versions of RBP2GO is that the Disease ontology information
search options now cover all species by default. Searching for RBP2GO 2.0 now provides users with the possibility to query
a specific species is possible by selecting the species of interest the database for one or more DO terms, allowing them to ex-
in the dedicated dropdown menu provided in the Advanced plore the possible impact of RBPs in various human diseases.
Search. To facilitate this process, a DO tree has been implemented in
Each protein in the result list can be individually selected. the Advanced Search module, enabling users to directly add
This triggers a switch to the specific protein page. Each pro- the desired term to their search (Fig. 3). Users can either nav-
tein is annotated with extensive information, which is now igate through the DO tree until they find the term of interest,
organized through a well-structured sidebar according to the or they can perform a keyword search and select terms from
category of information and underlying datasets. The pro- a proposed matching list. The terms are then added to the
tein page is further expanded through links to external re- search field upon selection by the users. We purposely only al-
6 Cantarella et al.

Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

Figure 4. Visualization of protein sequence features. The position of intrinsically disordered regions, protein domains and families, and RNA-binding
peptides (XL-peptides) is accessible under the menu “Features & XL-Peptides” of the sidebar in the blue panel on the left. In the example of the
Heterogeneous nuclear ribonucleoprotein U (Homo sapiens), colored boxes represent domains or families identified as RNA-binding (grey parts are not
RNA binding or the information is not available/unknown). Additional details can be accessed by clicking directly on the boxes representing domains,
families, or XL-peptides. XL-peptides are grouped according to the studies in which they were identified, and a complete dataset from all studies can be
retrieved using the “Download All Peptide Datasets” button. For each individual dataset, the download button “Download Dataset” can be used.

low the DO search using specific DO terms to avoid confusion can be useful to select proteins of interest for further investi-
in determining the relation between RBPs and diseases. In to- gations.
tal, DO terms have been included in the database. Moreover,
the degree of association between each protein and a specific Sequence features and RNA-binding peptides
disease is quantified by a confidence score [30], which ranges For each protein with RNA-binding peptide information
from 0 to 5, 3 being considered as confident and >4 as highly (crosslinked peptides or XL-peptides), the positions of protein
confident. domains along the protein sequence are provided in addition
to general sequence feature information and the correspond-
ing InterPro IDs. IDRs, protein domains, and family informa-
Visualization of the RBP2GO Score and the tion are easily accessible under the “Features & XL-Peptides”
RBP2GO Composite Scores sidebar menu (Fig. 4). Detailed information, such as the start
In our previous publications, we computed two scoring sys- and stop position, is displayed by clicking on the desired fea-
tems that reflect the propensity of a protein to bind to RNA, ture. For each InterPro ID, a link to the corresponding InterPro
i.e. to be a true RBP [23, 26]. The RBP2GO Score is based page is provided for in-depth documentation.
on the frequency with which a protein and its binding part- One of the distinguishing features of RBP2GO 2.0 is the
ners are experimentally detected as RNA binding, and the inclusion of XL-peptide data along with InterPro domain and
RBP2GO Composite Score additionally integrates informa- family information (Fig. 4). The XL-peptide data were col-
tion on the presence of an RBD in the protein sequence or an lected using both experimental and in silico methods [35–40].
association with an RNA-related protein family. Importantly, Specifically, the datasets from Bae et al. (2020) [39] and Bae
the 15 new RBDs added since the RBP2GO-2-Beta version et al. (2021) [40] report only the positions of the crosslinked
have been incorporated into the calculation of the RBP2GO amino acids. To determine the associated XL-peptides, a pre-
Composite Score. For each protein, the RBP2GO Score and vious study had randomly extended the sequence surrounding
RBP2GO Composite Score are marked within the distribution the crosslinked residues [46]. Here, we computed the theoreti-
of scores of either all RBPs or all proteins of the same species, cal tryptic peptides by identifying the closest trypsin digestion
both depicted as violin plots. This new feature helps users de- sites—since trypsin was used in the aforementioned studies to
termine if the protein of interest has a higher tendency to bind generate XL-peptides and to further identify the crosslinked
to RNA as compared to other proteins of the same species and residues.
 Integrating disease associations and sequence features 7

XL-peptide data are easily accessible by clicking a spe- Author contributions: M.C.H. conceived the RBP2GO up-
cific feature, where the start and end positions, as well as the date project and supervised the project with the help of S.C.
amino acid sequence and corresponding crosslinked residues S.C. wrote the manuscript and prepared the figures. J.N.
(if available), are clearly provided. The inclusion of XL- computed the RBP2GO 2.0 version based on the previous
peptide data represents a valuable enhancement, allowing RBP2GO versions and integrated datasets prepared by M.H.,
users to collect information dispersed across multiple stud- F.R., R.S., P.J., and E.S. All authors have seen and approved

Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

ies and to facilitate the identification and validation of RNA- the final version of the manuscript.
binding regions within proteins of interest.

Conflict of interest
Conclusions and future perspectives The authors disclose no conflicts of interest.
Here, we introduce RBP2GO 2.0, an updated version of
our previously published database [26] and its online de-
velopment RBP2GO-2-Beta [23]. The database has been ex- Funding
panded through the addition of 19 new studies comprising Research on RNA–protein complexes in our lab is supported
30 datasets, resulting in a substantial increase in documenta- by DKFZ Core Funding and grants from the Dr. Rolf M.
tion on RBPs. Furthermore, data accessibility has been greatly Schwiete Stiftung and the DKFZ Hector Cancer Research In-
enhanced through the design of a new user interface, which stitute at University Medicine in Mannheim, Germany. Fund-
provides direct access to four different search boxes from the ing to pay the Open Access publication charges for this article
Home page, including a completely new search dedicated to was provided by DKFZ Core Funding.
the disease ontology. The new Home page represents one of
the key novelties of the RBP2GO database, allowing users to
quickly target the RBP searches according to their specific in- Data availability
terest. They can rapidly generate lists of proteins associated RBP2GO 2.0 is freely available at [Link]
with particular features and benefit from multiple download without registration. The datasets as well as the scripts of the
options. database (as Shiny App) are freely available under the “Refer-
Another novelty of our new version of RBP2GO is the vi- ences & Data” top menu. RBP2GO is optimized to work with
sualization of the RBP2GO scoring systems as violin plots, internet browsers such as Google Chrome and Safari.
which allows a fast and precise positioning of each protein in
the scoring landscape of each species. Importantly, the ability
to filter searches for RBP aspirants (non-RBPs with RBDs) us- References
ing the Advanced Search tool, combined with sorting results 1. Kishore S, Luber S, Zavolan M. Deciphering the role of
by descending scores, facilitates the identification of poten- RNA-binding proteins in the post-transcriptional control of gene
tial new RBPs. Lastly, the positions of RNA-binding peptides expression. Briefings in Functional Genomics 2010;9:391–404.
along the protein sequence are now easily accessible, provid- [Link]
ing additional and important information that, to our knowl- 2. Gerstberger S, Hafner M, Tuschl T. A census of human
edge, is not available in any other functional RBP database. RNA-binding proteins. Nat Rev Genet 2014;15:829–45.
The RBP2GO database will continue to evolve in the fu- [Link]
3. Van Nostrand EL, Pratt GA, Yee BA et al. Principles of RNA
ture. For example, the RBP homolog and ortholog datasets
processing from analysis of enhanced CLIP maps for 150 RNA
will be expanded to include resources such as The Alliance of binding proteins. Genome Biol 2020;21:90.
Genome Resources [47] in addition to the UniRef50 dataset [Link] 020- 01982- 9
[42] that is already in use. Future improvements to RBP2GO 4. Choi Y, Um B, Na Y et al. Time-resolved profiling of RNA binding
2.0 could also include graphical representations of database proteins throughout the mRNA life cycle. Mol Cell
statistics (e.g. bar plots of RBPs and non-RBPs across species) 2024;84:1764–1782.e10.
and flowcharts for suggested Advanced Search option com- [Link]
binations, useful, for example, to explore RBP enrichment in 5. Street LA, Rothamel KL, Brannan KW et al. Large-scale map of
specific diseases. Lastly, the accessibility of information could RNA-binding protein interactomes across the mRNA life cycle.
be further expanded by including links to the RCSB PDB Mol Cell 2024;84:3790–3809.e8.
[Link]
([Link] pages of protein structures in complex
6. Choudhury NR, Heikel G, Trubitsyna M et al. RNA-binding
with RNA. activity of TRIM25 is mediated by its PRY/SPRY domain and is
In the context of the emerging concept of “riboregulation” required for ubiquitination. BMC Biol 2017;15:105.
of non-canonical RBPs [7–11], we anticipate that RBP2GO [Link] 017- 0444- 9
2.0 will be a valuable resource to support the discovery of un- 7. Horos R, Buscher M, Kleinendorst R et al. The Small Non-coding
derexplored and unexpected RNA-RBP functions and to in- Vault RNA1-1 Acts as a Riboregulator of Autophagy. Cell
vestigate their associations with human diseases. 2019;176:1054–1067.e12
[Link]
8. Huppertz I, Perez-Perri JI, Mantas P et al. Riboregulation of
Acknowledgements Enolase 1 activity controls glycolysis and embryonic stem cell
differentiation. Mol Cell 2022;82:2666–80.
The authors sincerely acknowledge the IT Core Facility of the [Link]
German Cancer Research Center (DKFZ) for their precious 9. Spizzichino S, Di Fonzo F, Marabelli C et al. Structure-based
support and for hosting the RBP2GO Shiny app, and deeply mechanism of riboregulation of the metabolic enzyme SHMT1.
thank colleagues, friends, and users for their constructive feed- Mol Cell 2024;84:2682–97.
back. [Link]
8 Cantarella et al.

10. Rajagopal V, Seiler J, Nasa I et al. An atlas of RNA-dependent 30. Grissa D, Junge A, Oprea TI et al. Diseases 2.0: a weekly updated
proteins in cell division reveals the riboregulation of mitotic database of disease-gene associations from text mining and data
protein-protein interactions. Nat Commun 2025;16:2325. integration. Database (Oxford) 2022;2022:baac019.
[Link] 025- 57671- 3 [Link]
11. Hentze MW, Sommerkamp P, Ravi V et al. Rethinking 31. Baron JA, Johnson CS, Schor MA et al. The DO-KB
RNA-binding proteins: riboregulation challenges prevailing views. Knowledgebase: a 20-year journey developing the disease open
Cell 2025;188:4811–27. science ecosystem. Nucleic Acids Res 2024;52:D1305–14.

Downloaded from [Link] by Sino-British College/USST user on 26 November 2025

[Link] [Link]
12. Lukong KE, Chang KW, Khandjian EW et al. RNA-binding 32. Blum M, Andreeva A, Florentino LC et al. InterPro: the protein
proteins in human genetic disease. Trends Genet 2008;24:416–25. sequence classification resource in 2025. Nucleic Acids Res
[Link] 2025;53:D444–56. [Link]
13. Gebauer F, Schwarzl T, Valcarcel J et al. RNA-binding proteins in 33. Buchbender A, Mutter H, Sutandy FXR et al. Improved library
human genetic disease. Nat Rev Genet 2021;22:185–98. preparation with the new iCLIP2 protocol. Methods
[Link] 020- 00302- y 2020;178:33–48. [Link]
14. Kelaini S, Chan C, Cornelius VA et al. RNA-binding proteins hold 34. Van Nostrand EL, Freese P, Pratt GA et al. A large-scale binding
key roles in function, dysfunction, and disease. Biology (Basel) and functional map of human RNA-binding proteins. Nature
2021;10:366. [Link] 2020;583:711–9. [Link] 020- 2077- 3
15. Popper B, Scheidt T, Schieweck R. RNA-binding protein 35. Castello A, Fischer B, Frese CK et al. Comprehensive identification
dysfunction in neurodegeneration. Essays Biochem of RNA-binding domains in human cells. Mol Cell
2021;65:975–86. [Link] 2016;63:696–710. [Link]
16. Hashimoto S, Kishimoto T. Roles of RNA-binding proteins in 36. Mullari M, Lyon D, Jensen LJ et al. Specifying RNA-binding
immune diseases and cancer. Semin Cancer Biol 2022;86:310–24. regions in proteins by peptide cross-linking and affinity
[Link] purification. J Proteome Res 2017;16:2762–72.
17. Lunde BM, Moore C, Varani G. RNA-binding proteins: modular [Link]
design for efficient function. Nat Rev Mol Cell Biol 37. Panhale A, Richter FM, Ramirez F et al. CAPRI enables
2007;8:479–90. [Link] comparison of evolutionarily conserved RNA interacting regions.
18. Corley M, Burns MC, Yeo GW. How RNA-binding proteins Nat Commun 2019;10:2682.
interact with RNA: molecules and mechanisms. Mol Cell [Link] 019- 10585- 3
2020;78:9–29. [Link] 38. Queiroz RML, Smith T, Villanueva E et al. Comprehensive
19. Clery A, Blatter M, Allain FH. RNA recognition motifs: boring? identification of RNA-protein interactions in any organism using
Not quite. Curr Opin Struct Biol 2008;18:290–8. orthogonal organic phase separation (OOPS). Nat Biotechnol
[Link] 2019;37:169–78. [Link] 018- 0001- 2
20. Valverde R, Edwards L, Regan L. Structure and function of KH 39. Bae JW, Kwon SC, Na Y et al. Chemical RNA digestion enables
domains. FEBS J 2008;275:2712–26. robust RNA-binding site mapping at single amino acid resolution.
[Link] Nat Struct Mol Biol 2020;27:678–82.
21. Linder P, Jankowsky E. From unwinding to clamping - the DEAD [Link] 020- 0436- 2
box RNA helicase family. Nat Rev Mol Cell Biol 2011;12:505–16. 40. Bae JW, Kim S, Kim VN et al. Photoactivatable ribonucleosides
[Link] mark base-specific RNA-binding sites. Nat Commun
22. Järvelin AI, Noerenberg M, Davis I et al. The new (dis)order in 2021;12:6026. [Link] 021- 26317- 5
RNA regulation. Cell Commun Signal 2016;14:9. 41. Piovesan D, Del Conte A, Mehdiabadi M et al. MOBIDB in 2025:
[Link] 016- 0132- 3 integrating ensemble properties and function annotations for
23. Wassmer E, Koppany G, Hermes M et al. Refining the pool of intrinsically disordered proteins. Nucleic Acids Res
RNA-binding domains advances the classification and prediction 2025;53:D495–503. [Link]
of RNA-binding proteins. Nucleic Acids Res 2024;52:7504–22. 42. Suzek BE, Wang Y, Huang H et al. UniRef clusters: a
[Link] comprehensive and scalable alternative for improving sequence
24. Liao JY, Yang B, Shi CP et al. RBPWorld for exploring functions similarity searches. Bioinformatics 2015;31:926–32.
and disease associations of RNA-binding proteins across species. [Link]
Nucleic Acids Res 2025;53:D220–32. 43. Caudron-Herger M, Rusin SF, Adamo ME et al. R-DeeP:
[Link] proteome-wide and quantitative identification of RNA-dependent
25. Giudice G, Sanchez-Cabo F, Torroja C et al. ATtRACT-a database proteins by density gradient ultracentrifugation. Mol Cell
of RNA-binding proteins and associated motifs. Database 2019;75:184–199.e10.
2016;2016:baw035. [Link] . [Link]
26. Caudron-Herger M, Jansen RE, Wassmer E et al. RBP2GO: a 44. Caudron-Herger M, Wassmer E, Nasa I et al. Identification,
comprehensive pan-species database on RNA-binding proteins, quantification and bioinformatic analysis of RNA-dependent
their interactions and functions. Nucleic Acids Res proteins by RNase treatment and density gradient
2021;49:D425–36. [Link] ultracentrifugation using R-DeeP. Nat Protoc 2020;15:1338–70.
27. UniProt C. UniProt: the Universal Protein Knowledgebase in 2025. [Link] 019- 0261- 4
Nucleic Acids Res 2025;53:D609–17. 45. Luo Y, Hitz BC, Gabdank I et al. New developments on the
[Link] Encyclopedia of DNA elements (ENCODE) data portal. Nucleic
28. Gene Ontology C, Aleksander SA, Balhoff J et al. The Gene Acids Res 2020;48:D882–9. [Link]
Ontology knowledgebase in 2023. Genetics 2023;224:iya031. 46. Chu LC, Christopoulou N, McCaughan H et al. pyRBDome: a
[Link] comprehensive computational platform for enhancing
29. Szklarczyk D, Nastou K, Koutrouli M et al. The STRING database RNA-binding proteome data. Life Sci Alliance
in 2025: protein networks with directionality of regulation. 2024;7:e202402787. [Link] .
Nucleic Acids Res 2025;53:D730–7. 47. Alliance of Genome Resources Consortium. Updates to the
[Link] Alliance of Genome Resources central infrastructure. Genetics
2024;227:iyae049. [Link]

Received: September 14, 2025. Revised: October 23, 2025. Accepted: October 23, 2025
© The Author(s) 2025. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License ([Link] which permits
non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact reprints@[Link] for reprints and
translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact
[Link]@[Link].