#uniprot — Public Fediverse posts

Do you want to know how to get more (or anything) out of the #uniprot #kg? I had claude gather my training materials and present it on the web in a way that is easier to engage with.

Would you like to have a look? https://sib-swiss.github.io/sparql-training/

I have migrated from @akademienl , which was a wonderful instance but unfortunately and understandably shutting down.

I hope to share here on genomics.social some of the work I do with #uniprot #rhea and #swisslipids. Specifically around keeping that data #FAIR and useable. I am most vocal about semantic web approaches and #KGs and separating data from the APIs that we just happen to build.

#UniGe Prof Leuba recounts the start of #UniProt with #Swiss-Prot in 1986 (where I work). How this is an example of making data understandable, and how quality is a work in progress. This is a requirement for all scientific fields. This is not a technical question: but of scientific quality and integrity. #UniGe has the duty to train the next generation of researchers to have these qualities.

My colleagues replied "if the SPARQL query fails ask Jerven ;)" which you can do to! Use https://www.uniprot.org/contact where I or one of the other members of the UniProt team will be happy to help you get the most out of #UniProt :)

Google-deepmind wrote some skills for using #UniProt and I really second the advice "If a complex search query fails, try to use SPARQL instead of giving up."

Specific #UniProt example https://sib-swiss.github.io/sparql-examples/examples/UniProt/125_CORDIS_EURIO_cited_project_results_in_uniprot.html

To our government officials: Please invest our tax money in science, not in stupid wars.

Let me explain... while Python is keeping busy my laptop doing number crunching, I thought I'd share with you part of process of updating GNU Health HIS genomic packages to the latest genes, symbols, human natural variants and their phenotypes / diseases for 2026. Quite a lot of work, for the betterment of the society, ad honorem.

So again to all in @EUCommission, instead of putting millions of euros in weapons and in the macabre business of war, invest in science, education and healthcare. Thank you.

#OpenScience #GNUHealth #GNU #bioinformatics #genomics #HUGO #UniProt #genetics #health #PublicHealth

My Domain: proteome-wide scanning of TMDs

https://quantixed.org/2026/04/16/my-domain-proteome-wide-scanning-of-tmds/

#science #bioinformatics #CellBiology #proteins #Rstats #Uniprot

Love how https://consensus.rubalkhali.science/ links ecological data to the molecular world of #UniProt with some smarts and the power of federated #SPARQL

@abrokenjester unfortunately is a changing measure. I still regret telling the @[email protected] that #UniProt will stay smaller tan 27 billion triples :( which is 1/10th of the current size.

@harmonia_amanda the vast majority of #SPARQL experts are autodidact! Heck I learn more from the #uniprot sparql endpoint users with their user cases than from the language spec. The joy of seeing smart scientist play with the data is amazing

https://www.europesays.com/ie/198966/ AlphaFold Database Structure Extractor: a web server and API to download AlphaFold structures using common protein accessions | BMC Bioinformatics #algorithms #AlphaFoldDB #APIToolkit #Bioinformatics #ComputationalBiology #ComputationalBiology/Bioinformatics #ComputerAppl.InLifeSciences #Éire #IE #Ireland #Microarrays #NCBI #OpenAccess #ProteinStructure #Science #StructuralGenomics #UniProt

User benefits of #uniprot are estimated to be on the order of 20 times higher than operational costs https://www.sib.swiss/news/uniprot-user-benefits-up-to-39-times-higher-than-operational-costs

#infrastructure #opendata #curation #FAIRdata #proteins

#UniProt The largest open+free #knowledgegraph in the life sciences. I will be talking about the why and the use cases at #semantics2025 5th of September 2025. Looking forward to seeing you all there!

https://2025-eu.semantics.cc/page/sessions?page=1nZsC6p2V3Yk73axR9Jl1NWq1IhVpkqtN

Interesting in his #ISMBECCB2025 keynote that Amos Bairoch described as the worst decision of his career: caving to USA pressure to switch from funding model of free to academics & non-profits to use #SwissProt / #UniProt (with no login barrier), but paid access to Industry - to a free-to-all model funded by an NIH grant.

[I suspect industry lobbying at NIH there. One issue with grants is they are typically fixed term. Also, this would have been a very strong precedent for other DBs to follow - would have been more sustainable ecosystem?]

#uniprot just hides the controls.. so you can't even try to full screen. but you can fortunately link out the #ebi #alphafold site where (ctd..)

🧬 How do AlphaFold2 and ESMFold stack up when it comes to functional annotation?

🔗 AlphaFold2 and ESMFold: A large-scale pairwise model comparison of human enzymes upon Pfam functional annotation. Computational and Structural Biotechnology Journal, DOI: https://doi.org/10.1016/j.csbj.2025.01.008

📚 CSBJ: https://www.csbj.org/

#AlphaFold2 #ESMFold #AIinScience #ProteinStructure #Enzymes #Pfam #UniProt #AlphaFold #HumanProteome #FunctionalAnnotation #AIinBiology #ComputationalBiology #FunctionalGenomics

In 2023, ISB members voted to award Sandra Orchard the Exceptional Contribution to Biocuration award (https://www.biocuration.org/pascale-gaudet-and-sandra-orchard-recipients-of-the-2023-exceptional-contribution-to-biocuration-award).

Today at #biocuration2025 in Kansas City, she's presenting her award talk "How will the role of the Biocurator change in this world of Artificial Intelligence and Machine-Learning?" with a focus on current work at UniProt

#biocuration #uniprot #mods #proteins #complexportal

At @swat4hcls there was a presentation by @albdrg that I really like: https://www.youtube.com/watch?v=F4Nl-nmLZA4 This uses #SPARQL to combine data in a privacy aware #beacons-api and combine it with public #knowledgegraph like #uniprot by @SIB and #wikidata @wikidata to reduce cost of the data integration challenge required to answer clinical question.

Slide from my talk "Open Science for Medicine and Society" at Komazawa university. Using GNU Health to select a BRCA1 variant involved in breast #cancer and the 3D representation of the BRCT domain using PyMOL.
Having access to the latest scientific evidence makes a huge difference in patients, yet medical research is being dismantled by #corporatocracy.
Medicine and society evolves not because of the bullshit hype of AI, but because of the effort and talent of the men and women behind these projects. Respect :anarchoheart3: 🩺
#UniProt #HUGO #GNUHealth #PyMOL #NIH #genomics #OpenScience

a bit over 17 years ago: "One Billion Biochemical RDF Triples!" https://chem-bla-ics.linkedchemistry.info/2007/10/24/one-billion-biochemical-rdf-triples.html

#semweb #uniprot

Very honored by the kind acknowledgements in https://academic.oup.com/nar/advance-article/doi/10.1093/nar/gkae959/7848838 by @ju I am very existed about doing cross database queries between #uniprot #rhea and media data in #bacdive @dsmz

If the protein or complex was purified from its native host organism:
1. I sequence the map with `model_angelo build_no_seq`.
2. I identify all proteins with `model_angelo hmm_search` against the reference proteome of the host organism (downloaded from #Uniprot).

If the protein or complex was prepared recombinantly, I already know the sequences of all proteins.

So at this stage, either way, I know which proteins are in there.
3/19

With we I actually mean: you the #UniProt user base ;)

#DidYouKnow that institutions are using #GNUHealth for the diagnostic and management of #genetic diseases? In addition, and thanks to our friends from #uniprot we provide the large human natural variants dataset & related conditions.
Check out the genetics chapter from our official documentation for more information about #genomics and medical genetics 👇
https://docs.gnuhealth.org/his/userguide/modules/genetics.html

Moving forward with #OpenScience ♥️ 🧬
#PrecisionMedicine

Looking forward to talk about #FAIR #UniProt and #SPARQL later today https://cost-dkg.eu/talks#UniProt%20and%20others:%20FAIRly%20large%20and%20connected%20SPARQL%20in%20the%20life-sciences

Missed the course on #Enzymes in #UniProt yesterday. Rest assured, the presentation will be available on the @SIB youtube channel soon (https://www.youtube.com/@sib-swissinstituteofbioinf4579). The course materials are already available at https://education.expasy.org/cours/SIB_UniProtKB_Enzymes_2024/

How #graph layout tools react to the #uniprot #rdf database schema 😉

I am looking for an #ontology #schema that I can use to mark up where #sparql example is known to work. e.g. a query that works for #uniprot and #bgee or #omabrowser. Please share for reach.

Talking about remodeling other people's data. There used to be #GeneOntology rdf published by #uniprot that looked nothing like what the GO published. We stopped doing that a long time ago (https://www.uniprot.org/release-notes/2015-03-04-release) and while GO in #OWL is not that easy to query in #SPARQL I can at least use the same queries internally and externally.

Looking at the next #uniprot release. Looks like the #uniprot #sparql endpoint grew roughly by half a wikidata (from 2023_05 to 2024_01). Check the numbers in 3 hours when 2024_01 should go live

@openlink

Woot!

Another update for the #VirtuosoRDBMS (Open Source Edition) is here! Keep in mind that all my live demonstrations, whether they use #SPARQL or #SQL to fine-tune #ChatGPT, are also compatible with this edition (for example, the 100 billion+ #Uniprot instance-related demos at: https://community.openlinksw.com/t/using-sparql-query-templates-to-fine-tune-chatgpts-large-language-model-llm/3842/5).

#RDF #GraphDatabase #SPARQL #LinkedData #SQL #GraphQL #SemanticWeb #KnowledgeGraph

**Problem:**
#LLM based #SmartAgent solutions like #ChatGPT are fantastic natural language processors, but challenged by unpredictable guessing.

**Solution:**
Fine-tune using #SPARQL query templates and a #LODCloud #KnowledgeGraph like @Uniprot.

https://community.openlinksw.com/t/using-sparql-query-templates-to-fine-tune-chatgpts-large-language-model-llm/3842/5#uniprot-knowledge-graph-chatgpt-fine-tuning-using-sparql-1

#SemanticWeb #LinkedData #HowTo #LLMs #Uniprot #Bioinformatics #Genomics #MolecularBiology #LifeSciences

The @fairsharing #UniProt graph also shows you how each of the resources linked to Uniprot connect **to each other**, which is a fantastic measure of the collaborative space around this resource. Go on, let us know, what is your favourite record with us?

(also shared on the bird site in modifed form)

Most interestingly to me, the record describes the ecosystem of databases, standards and policies connected to #UniProt, visualised by the graph I shared in my first toot. The colours of the graph edges tell you a lot. Here's the legend, for reference.
What does its @fairsharing graph tell you about #UniProt? The multitude of orange says it's listed in many policies. The large pink region says it implements many standards. Note that its graph also shows another kind of interconnectedness..

#SIB Preprint on the power and limitations of #AI #chatbots such as #chatgpt to understand #SPARQL federated queries over #UniProt, #OMA and #Bgee https://arxiv.org/abs/2304.10427

For many years I've observed the following in parallel:

1. General confusion (and in some cases disdain) about the notion of a #SemanticWeb

2. Steadfast effort by a community to keep the vision going, stealthily, as exemplified by the #LODCloud (#DBpedia,, #Wikidata, #Uniprot etc..) and Schema.org (#SchemaOrg)

3. #DataSilos and all the impedance they inflict on the agility of individuals and enterprises alike on the rise i.e., #DataAccess and #DataFlow are still challenging, unnecessarily.