Omid Arhami

GitHub - omid-arhami/topolow: A physics-inspired R package for robust Euclidean embedding of sparse, non-metric dissimilarity data. Particularly powerful for antigenic cartography and viral evolution ... A physics-inspired R package for robust Euclidean embedding of sparse, non-metric dissimilarity data. Particularly powerful for antigenic cartography and viral evolution tracking, but applicable to...

Please download version 2 from GitHub to use the new function Euclidify until September, when CRAN resumes publishing updates.
github.com/omid-arhami/...

We explained and proved the theories of Topolow, as an Euclidean embedding algorithm for general data science. We showed it has significantly greater accuracy at transforming non-metric data to metric, a crucial step before using most statistics and machine learning tools.
github.com/omid-arhami/...

The way we name our #manuscript revisions on @overleaf.com

We tested Topolow on the classic H3N2 influenza data. Our map is similar to the lab-validated map AND our antigenic velocity vectors automatically identify cluster transitions previously found with reverse-genetics.
See analyses in our Bioinformatics paper academic.oup.com/bioinformati...
#IDSky 🧪

Thrilled to announce our new paper in Bioinformatics. We present Topolow, a physics-inspired algorithm to create more accurate & stable antigenic maps, even from sparse data; also "antigenic velocity" showing the direction and magnitude of phenotypic changes.
🔗Paper: academic.oup.com/bioinformati...

#Tehran metro is 863 sq miles, all densly populated. It's twice the size of the entire Rocky Mt park.
All 5 boroughs of NYC are 303 sq mi.

Avian influenza A viruses exhibit plasticity in sialylglycoconjugate receptor usage in human lung cells | Journal of Virology It is well known that influenza A viruses (IAV) initiate host cell infection by binding to sialic acid, a sugar molecule present at the ends of various sugar chains called glycoconjugates. These sugar...

(7/7) So, it's not just about the linkage type (α2,6 vs α2,3).
journals.asm.org/doi/10.1128/...

(6/7) efficiently, revealing they could use alternative, less common glycoconjugates like sialyl Tn and sialyl T antigens.
This adaptability in receptor usage may explain why avian influenza viruses can more readily jump between species.

(5/7) C1GALT1, and UGCG
The most striking discovery was that when all three glycoconjugate types were truncated:
1.Human IAV strains showed a dramatic 1-3 log decrease in replication, suggesting they require at least one of these major glycoconjugate types.
2.Avian IAV strains continued replicating

(4/7) The researchers used #CRISPR #gene editing on human lung cells, naturally expressing both α2,6-linked and α2,3-linked sialic acids distributed across all three glycoconjugate types, to create lung cells lacking different combinations of glycoconjugate types by knocking out key enzymes: MGAT1,

(3/7) The paper examines how #sialic acids can be attached to three different types of molecules on cell surfaces (#glycoconjugates):
1- N-glycans
2- O-glycans
3- Glycolipids
No one had systematically tested how different IAV strains might differ in their use of these three glycoconjugate types.

(2/7)
- Human IAVs prefer sialic acids with α2,6 linkages
- Avian IAVs prefer sialic acids with α2,3 linkages
But this collaboration by researchers from University of Iowa, U. of Georgia, U. of Chicago, U. at Buffalo, and UCSD reveals that the story is more complex than just the type of linkage.

(1/7) A fascinating editor's pick in the Journal of #Virology reveals why #avian influenza viruses may be particularly adept at jumping between species. Influenza A virus (IAV) has two groups of strains: avian IAV and human IAV; the key difference has traditionally been understood as:

66 | Michael Fischbach and the scientific decision tree Night Science · Episode

The Night Science podcast is fantastic. I enjoyed each individual episode of it I listened to.
open.spotify.com/episode/6KMh...

Right. But one example doesn't negate a whole concept; It shows that peer review isn't 100% error-proof. But it's still way better than no control.

🥹

Peer-reviewed journals are like municipal courts. They cost money but we can't get rid of them and let people on streets decide who is a criminal and who is not. I think the solution is to regulate the profit journals demand with something like fare-trade rules.

www.truefood.tech
A large, but not exhaustive, database of foods scored by the amount of additives.
It doesn't show how nutritious, protein-rich, healthy,... a food is.
It's not perfect, but good for beginners.

Such an invigorating piece of news 🙂 When I was testing a new #data analysis method to potentially contribute to #vaccine development for #HIV viruses, I accidentally saw this news. #science

The glycosyltransferase ST3GAL4 drives immune evasion in acute myeloid leukemia by synthesizing ligands for the glyco-immune checkpoint receptor Siglec-9 - Leukemia Leukemia - The glycosyltransferase ST3GAL4 drives immune evasion in acute myeloid leukemia by synthesizing ligands for the glyco-immune checkpoint receptor Siglec-9

A new paper by people I admire on ST3GAL4 driving #immune evasion in acute myeloid #leukemia by synthesizing #ligands for Siglec-9. 🧪 #science @natureportfolio.bsky.social
www.nature.com/articles/s41...

Pej Rohani on LinkedIn: A single mutation in bovine influenza H5N1 hemagglutinin switches… Epidemiologists are perennially worried about viruses jumping the "species barrier" from animal species to humans and causing a(nother) pandemic. For avian…

www.linkedin.com/posts/pej-ro...

Scientists, academics, researchers: We’re excited to share that @altmetric.com is now tracking mentions of your research on Bluesky! 🧪