rfam / rfam-website Goto Github PK

Currently Rfam families are associated with GO terms but no additional metadata is captured. It would be useful to provide the following for all GO terms:

• Qualifier (especially important if it is 'NOT’)
• Reference (e.g. PubMed ID or a GO_REF describing how the annotations were made)
• Evidence code
• Date
• Assigned by

Documentation: ftp://ftp.ebi.ac.uk/pub/databases/GO/goa/HUMAN/README

If implemented, the information can be propagated to RNAcentral.

• As many families as possible need to have GO terms.
• GO terms need to be consistent within RNA types (for example, all riboswitches should have a translation regulation GO term).

I tried to download a fasta alignment of motif and I got an empty file in safari. For example the fasta alignment here http://rfam.xfam.org/motif/RM00004#tabview=tab1 gives an empty file. The stockholm file looks correct.

Nagios needs a page with a standard string of characters that will only be present if the site is up and the database is available.

We need to add Genome as a fourth entity type in Rfam search in addition to Family, Clan, and Motif.

The Genome object should contain the following fields:

• uniprot_reference_proteome_id - empty if UPID not available
• rfam_genome_id - empty if UPID is available
• gca_accession - empty if GCA not available
• description
• length - in nucleotides
• taxonomy_lineage - string like Bacteria; Firmicutes; etc
• ncbi_taxonomy_id cross reference - see comment below
• num_rfam_hits - number of significant Rfam family hits
• num_rfam_families - number of distinct Rfam families with significant hits

The flavivirus families have an error on the Summary web page, the wiki info is not loading.
These families have an auto_wiki identifier of 2759 which is Flavivirus_3_UTR - however, https://en.wikipedia.org/wiki/Flavivirus_3_UTR is not a valid entry.
It seems these families should be 2754 Flavivirus_3'_UTR
Will change these families to Flavivirus_3'_UTR by updating their auto_wiki ID in the database.

RF03536
RF03537
RF03538
RF03539
RF03540
RF03541
RF03542
RF03544
RF03545
RF03547

• add a facet has_pseudoknot to find families with pseudoknots
• add export in CSV/JSON formats + CM download option
• index more data (number of non-gapped columns, alignment length etc)
• add autocomplete
• index GO/SO term definitions, not just ids
• add alphabetic sorting by name option
• hide 0 sequences, 0 structure, 0 species brown panel at the top of the search results page
• index all taxa (Mammalia, Bacteria etc), not just the terminal leaves
• hide text labels in R-scape previews
• show an alternative 2D preview in search results when R-scape SVG is not available (e.g. RF01795)

It seems that the file containing mapping of rfam families to pdb codes is incomplete in rfam 14.0 and 14.1.

In 14.0 (ftp://ftp.ebi.ac.uk/pub/databases/Rfam/14.0/Rfam.tar.gz) pdb code is missing:

pdb_id	chain	pdb_start	pdb_end	bit_score	evalue_score	cm_start	cm_end	hex_colour
RF00001	C	3	118	77.20	1.6e-20	1	119	0064f4
RF00001	9	1	121	77.40	1.4e-20	1	119	ebeb30
RF00001	9	1	121	77.40	1.4e-20	1	119	93c090
RF00001	B	1	121	77.40	1.4e-20	1	119	c008ae
RF00001	B	1	121	77.40	1.4e-20	1	119	8484c0

In 14.1 (ftp://ftp.ebi.ac.uk/pub/databases/Rfam/14.1/Rfam.pdb.gz) the header is missing and there is no rfam ids:

3hjw	D	1	57	48.50	1.3e-11	1	70	1fc01f
3lwo	D	1	57	48.50	1.3e-11	1	70	ff87a4
3lwp	D	1	57	48.50	1.3e-11	1	70	ebeb30
3lwq	D	1	57	48.50	1.3e-11	1	70	f29242
3lwr	D	1	57	48.50	1.3e-11	1	70	8585e6

• provide downloadable files on FTP
• add links to sequence and genome summary pages

Request to bring back the average sequence length field, which will allow sorting through families based on increasing length. User feed back: It was very useful having an overview of sequence lengths in Rfam families when selecting a data set.

Add a page that lets the users browse GO and SO terms associated with Rfam families.

It looks like some PDB chains do not match the right family.

For example, chains 1A and 2A from 5DOX should match the Bacterial LSU but they match the 23S pseudoknot (RF01118).

Trying to list all the Rfam families that have pseudoknots and 3D structures, but at the moment it is not possible to sort by the # of 3D structures and would be very useful.

Blame @kalvari

Example: http://www.sanger.ac.uk/cgi-bin/Rfam/getacc?RF00177

There is a new version of R-chie - we should try to install it and switch to it on the Rfam website soon:
https://academic.oup.com/nar/advance-article/doi/10.1093/nar/gkaa708/5911741

• indicate how many Rscape-significant base pairs are not shown in the displayed structure. This would be both a quality control hint (if an Rfam curated structure was messed up and missing a bunch of base pairs), plus an indication of whether a pseudoknot might be present.

• include a note of how many Rfam-annotated base pairs are not shown in the Rfam structure because they're an annotated pseudoknot.

• display pseudoknots. Perhaps draw a bracket over a contiguous string of nucleotides that are one side of the PK stem and lettering the two halves A and a, B and b.

The left sidebar links are broken when the site is loaded over https.

NDB is being replaced by NAKB (https://nakb.org/). The links in Rfam that point to NDB should be replaced by links to NAKB.

For example, GNRA Rfam motif links to Tetraloop on Wikipedia but the wiki content does not appear on the Rfam page.

I used web-search to search for accession AP011496.1 (RF02547 : mtPerm-5S) and the sequence matched all the families in the LSU clan apart from the 5S RNA family I took the sequence from.

The families are listed below:
RF02543 - LSU_rRNA_eukarya
RF02541 - LSU_rRNA_bacteria
RF02540 - LSU_rRNA_archaea

The website had a bug where CMs could not be downloaded. It isn't clear why but this should be fixed. We should check it post release 14.9.

• num_full field in the family table is not updated
• no sequences are marked as seed in the Sequences tab
• PDB chains are still redundant (example: chain 0 from 1s72)
• Schistosoma japonicum is not found in RF00163 anymore - no reference proteome for this species
• delete sRNA type in RF01684 so that it's just Gene
• check that families from this search do not have 0 species http://rfamlive.xfam.org/search?q=GO:0039703
• species sunburst has Unknown/Uncategorised entries (example: SAM)
• select * from family where number_of_species = 0; should not return any rows
• fix all families with 0 species - example: RF00162
• Example where the sunburst does not include all species (Pseudomonas syringae pv. tomato str. DC3000 is not in the sunburst) http://rfamlive.xfam.org/search?q=RF02749
• add seed rfamseq entries so that sunbursts have the same species as trees

Each author name should appear on a separate line in DESC files followed by ORCID id:

AU Author AB 0000-0000-0000-0000

• add ORCID to curation tools
• display on the website

Mapping spreadsheet

It seems not all HTML alignments can be viewed. One which cannot is:

https://rfam.org/family/RF04222#tabview=tab2

When trying to view the HTML alignment nothing shows up.

Rfam Curator needs to be able to add or delete a genome using UPID or GCA accession.

The websearch currently does not clan compete hits resulting in spurious results.

Use is_significant field to filter the entries.

• update header and footer
• show featured families
• embed Twitter feed
• remove old search options

Fold the sequence using the CM, show the structure in Rfam and export it to RNAcentral too.

Several users pointed out that it's difficult to find where to download various family specific files, so I suggest creating a new Downloads tab on the left hand side of Rfam family pages. It would aggregate all the downloadable items that are spread across all the other tabs, such as:

• Covariance model
• All FASTA sequences
• Sequences as a table
• Seed alignment in various formats
• Seed tree
• Species tree
• Secondary structure diagrams as images

There may be other downloadable items that I missed so please click around the family pages to double check.

We can keep the existing download links as they are or drop them in favour of the new tab (this can be decided on a case by case basis).

RNA Thermometer - SO:0002168

See The-Sequence-Ontology/SO-Ontologies#404 for details.

Hi,

This fails with Ubuntu 20.04:

curl https://rfam.xfam.org
#curl: (35) error:14094410:SSL routines:ssl3_read_bytes:sslv3 alert handshake failure

but works fine with Ubuntu < 20.04 and on Windows and macOS Mojave.

This seems to happen with some websites because of a combination of three reasons: server misconfiguration, increased TLS security level in Ubuntu 20.04 by default, and a bug in OpenSSL 1.1.1. See Ensembl/ensembl-rest#427 for a similar issue with the Ensembl server.

FWIW this breaks Bioconductor package rfaRm: https://bioconductor.org/checkResults/3.12/bioc-LATEST/rfaRm/nebbiolo1-install.html

Internally the package tries to access rfam.xfam.org with the following R code:

> library(xml2)

> read_xml("https://rfam.xfam.org/clans")
Error in open.connection(x, "rb") :
  error:14094410:SSL routines:ssl3_read_bytes:sslv3 alert handshake failure

> sessionInfo()
R version 4.0.2 Patched (2020-08-04 r78971)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04.1 LTS

Matrix products: default
BLAS:   /home/hpages/R/R-4.0.r78971/lib/libRblas.so
LAPACK: /home/hpages/R/R-4.0.r78971/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] xml2_1.3.2

loaded via a namespace (and not attached):
[1] compiler_4.0.2 curl_4.3      

Thanks!
H.

Example

Species sunburst for Clostridia in RF01315 shows that there are 64 sequences:

An example SQL query confirming the number of sequences:

SELECT CONCAT(t1.rfamseq_acc, '/', seq_start, '-', `seq_end`)
FROM full_region t1, rfamseq t2, taxonomy t3
WHERE t1.rfam_acc = 'RF01315' 
AND t1.rfamseq_acc = t2.rfamseq_acc
AND t2.ncbi_id = t3.ncbi_id
AND t3.tax_string LIKE '%Clostridia;%'
AND is_significant = 1
GROUP BY rfamseq_acc;

64 rows (like in sunburst UI) - note the GROUP BY clause

However, there are many more annotated regions:

SELECT CONCAT(t1.rfamseq_acc, '/', seq_start, '-', `seq_end`)
FROM full_region t1, rfamseq t2, taxonomy t3
WHERE t1.rfam_acc = 'RF01315' 
AND t1.rfamseq_acc = t2.rfamseq_acc
AND t2.ncbi_id = t3.ncbi_id
AND t3.tax_string LIKE '%Clostridia;%'
AND is_significant = 1;

6222 rows - no GROUP BY clause

So the number of entries in the resulting FASTA file is inconsistent with sunburst UI.

When a user is on an Rfam page and clicks on one of the tabs on the left panel, this action redirects him/her to the Rfam home page rather than displaying the requested data on the page.

This issue only occurs when accessing Rfam using the old URL http://rfam.xfam.org. Website functionality through the new URL http://rfam.org seems normal.

Retire old URL?

Sequence download is not working correctly. The sequence extracted isn't the correct one. For example there is an inconsistency between the SEED sequence of family RF00005

AM286415.1/4382210-4382145
GAAGGCACGACAUUGCUCACAUUGCUUCCAGUGUUUACUU-AGCCAGC---CGGG-UGCUGGCUUUUUUUU

and the sequence downloaded directly from the website:

AM286415.1/4382210-4382145
CTTCCGTGCTGTAACGAGTGTAACGAAGGTCACAAATGAATCGGTCGGCCCACGACCGAAAAAAAA

When a user submits a job using Job Dispatcher APIs, a job identifier (i.e. iprscan5-R20170926-112045-0446-74958378-es) is returned that will also include "-p1m" and "-p2m" at the end of the string. This is in addition to current values "-es", "-pg" and "-oy". The latter three will eventually disappear and another announcement will follow with proposed dates. These characters denote independent job clusters, which we use to provide resilience and failover.

If the code takes into account these last characters after the dash please take note of the change.

Examples of existing job identifiers:

iprscan5-R20170926-112045-0446-74958378-es
ncbiblast-R20170926-112324-0724-35376033-pg
emboss_needle-I20170926-112309-0557-5534481-oy

In addition to the above you should expect:

emboss_needle-I20170926-112309-0557-5534481-p2m
ncbiblast-R20170926-112324-0724-35376033-p1m
emboss_needle-I20170926-112309-0557-5534481-p2m