geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
223 stars 40 forks source link

RNA chaperone? #22499

Closed RLovering closed 2 years ago

RLovering commented 2 years ago

Following from MF SLIM project Val noted that: GO:0140691 RNA folding chaperone has only a single annotation in the entire GO. Maybe RNA binding would suffice for now?

RLovering commented 2 years ago

I don't know enough about RNA chaperones or RNA folding to know how many proteins are involved in the process. However, I am concerned that unless there has been an annotation project focused on this area it is difficult to be sure whether this term is required. In addition, following from our recent MF SLIM discussions where the 'binding' domain of GO was considered to be excluded, and then included I have realised that more MF terms are required for RNA functions. This article https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC6086601/ suggests that the MF term should stay but that an annotation project should focus on populating this term with relevant gene products.

RLovering commented 2 years ago

Maybe we should contact this group https://www.phd-rna-biology.at/network/ and although this is a 2007 article perhaps ask them for a list of RNA chaperones based on this https://www.tandfonline.com/doi/pdf/10.4161/rna.4.3.5445?needAccess=true (the website list in this article doesn't work www.projects.mfpl.ac.at/rnachaperones). However, this brings in the idea that RNA chaperones have a variety of activities some of which I know you would want to classify as processes ;( See Fig 1 (this is about the assays but I think these descriptions are useful): A) RNA chaperone activity (RCA) on double-stranded RNA can result in:

helix destabilization (duplex unwinding)
full strand dissociation (RNA melting)
strand displacement (strand exchange)
(B and E) Cis-splicing: Proteins with RCA significantly increase the population of pre-RNA molecules with splicing competent conformation.
(C) Trans-splicing: if the pre-RNA is transcribed in two pieces, which have to anneal for the formation of an active ribozyme. Proteins with RCA facilitate trans-splicing by bringing 2 RNAs together appropriately.
(D) RCA may resolve misfolded hammerhead ribozyme complexes (requires two activities—annealing and strand displacement).
(F) RCA may resolve misfolded to prevent early transcription termination.

The article then defines specific protein families:

Heterogeneous nuclear ribonucleoproteins (hnRNPs).

hnRNP I and hnRNP K are active in splicing assays
hnRNP A1 was demonstrated to enhance hammerhead ribozyme cleavage.
hnRNP A1, hnRNP C1/C2, hnRNP U, hnRNP K and hnRNP I promote RNA annealing.

Cold shock domain proteins (CSPs) (9 CSPs in E. coli)

CspA and CspE function as transcription antiterminators or translational enhancers at low temperature, where RNA secondary structures become inhibitory.
ribosomal protein S1 has RNA chaperone activity in the context of ribosomal proteins.
translation initiation factor 1 (IF1), has RNA chaperone activity: splicing of a misfolded intron and resolves terminator stem.
in eukaryotes: YB-1, which is a general regulator of translation by modulating mRNA structural rearrangements and packaging. promotes non-specific strand displacement and annealing of both RNA and DNA.

Ribosomal proteins.

RCA activity suggested for S1, S12, L1, L3, L13, L15, L16, L18, L19, L22 and L24

Histone-like proteins from bacteria.
Viral nucleocapsid proteins (NCps)

Then classifies these according to 'function': RNA Annealers DEXD/H-BOX (RNA heliicases) plus RCAs intrinsically unstructured proteins

Interesting statements from another article: https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC7404189/

cotranscriptional RNA folding constitutes a dynamic and intricate process so Colin maybe able to shed light on the RNA chaperones at the transcriptional level.
RNA Modifications, Editing and Sequence Composition as affecting the RNA structure.
Helicases comprise the largest group of transient remodelers of DNA and RNA structural arrangement: Some of them translocate along an RNA strand, unwind RNA duplexes, and displace proteins, while others are capable of solely unwinding dsRNA regions and/or mediating RNA-annealing
RBPs exert local RNA structural alterations as well as changes of structural context of adjacent regions due to torsional stress and thermodynamic compensation of local alterations.
I think Table 1 https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC7404189/table/ijms-21-05161-t001/ might provide a list of RBPs that modulate RNA structure.

Note that MF term GO:0008248 obsolete pre-mRNA splicing factor activity removed because this is a process. Of the 353 human proteins associated with the GO term GO:0008380 RNA splicing only 107 have MF annotations that are not a binding term. I have got the spreadsheet I used for this. But looking at just the first one I noticed without an MF (that was not a binding term) is https://www.uniprot.org/uniprot/P22626 prthe UniProt summary states: The hnRNP particle arrangement on nascent hnRNA is non-random and sequence-dependent and serves to condense and stabilize the transcripts and minimize tangling and knotting. Not sure which of the proteins in the hnRNP are responsible for the RNA chaperone activity but presumably some are.

RLovering commented 2 years ago

RNA_chaperone.xlsx

pgaudet commented 2 years ago

Following from MF SLIM project Val noted that: GO:0140691 RNA folding chaperone has only a single annotation in the entire GO. Maybe RNA binding would suffice for now?

I took this to mean, "Maybe RNA binding would suffice for now?", not in GO. This term was created recently, we dont expect it would have many annotations.

@ValWood I am correctly interpreting what you wrote?

ValWood commented 2 years ago

I was suggesting not to include 'RNA chaperone' in the slim until it was used, but ignore. It seems that a lot of RNA binding annotations will begin to migrate to chaperone.

pgaudet commented 2 years ago

Removed from both subsets - generic & prok