👾 Karthik

Got milk virus genomes?  (or any H5N1?)  Throw them at usher.bio and see where they land in per-segment trees of tens of thousands of public Influenza A H5N1 sequences!

For a report on our analyses of #H5N1 in cattle, see worobey bsky social

My colleagues and I have just posted a "Note about availability of #H5N1 2.3.4.4b consensus sequences from cattle and other species" @virological_org. Consensus genomes and metadata. virological.org/t/note-about-a…

#GPUalgorithms achieve significant speedup in statistical #phylogenetics, exemplified by estimating West Nile virus introduction timing from 104 genomes using BEAGLE v4.0.0 in BEAST. [co-authored by EVBC members @LemeyLab @suchard_group] doi.org/10.1093/bioinf…

@ScienceStanley @DiseaseOutbreak @GISAID Glad you find it useful!

@gkay92 @DiseaseOutbreak @GISAID This is so amazing!

We've been working hard to enable quicker and easier ⚡️ access to simple but key SC2 variant statistics 📉📈 on @DiseaseOutbreak Our R package now lets any user with @GISAID credentials to quickly access these stats for surveillance efforts 🔍 & complex downstream analyses 🤓

In a @VanityFair story, @KatherineEban oddly misrepresented an investigation into @GISAID and Peter Bogner by @sciencecohen and me; suggested our story was a volley in the lab leak debate; and let Bogner dismiss it as a "hit piece." Here's our response. science.org/do/10.1126/sci…

My colleague @sciencecohen and I spent months investigating @GISAID and the mysterious man behind it, Peter Bogner. Here's what we found. science.org/content/articl…

Great to see these API connections!

🧵for our latest preprint on the intrahost evolution of SARS-CoV-2 virus in an immunocompromised individual (60s) with a history of cancer chronically infected for at least 471 days (ongoing) with consistently replicating viruses at a high viral load. 1/n medrxiv.org/content/10.110…

@TWenseleers @RobLanfear @flaneuseks @K_G_Andersen I’m not aware of such an R package but might be worth connecting with GISAID regarding their api feeds through which they provide programmatic access.

@gkay92 @RobLanfear @flaneuseks @K_G_Andersen Thx! And would you perhaps know of an R package that would allow one to programmatically get those downloads from GISAID, by mimicking what one would do if one manually entered credentials and clicked on those downloads?

@TWenseleers @RobLanfear @flaneuseks @K_G_Andersen Sorry but the API can only be used to access the aggregated metrics on our database. The direct downloads are accessible only via GISAID.

@gkay92 @RobLanfear @flaneuseks @K_G_Andersen Is there any way to get the complete GISAID metadata or GISAID genomic epidemiology metadata download through that API by any chance? Wondering how to get those 2 downloads in R programmatically given one's credentials (only know how to get GISAID JSON from R)?

@RobLanfear @flaneuseks @K_G_Andersen We have vignettes but we should document this usage for sure 😅

@flaneuseks @RobLanfear @K_G_Andersen In the meantime, we actually have this endpoint that can be access in a hacky way thru R pkg. url <- "api.outbreak.info/genomics/preva…" df <- getGenomicsResponse(url)[[1]]

@malkuzweny @GISAID @TAlexPerkins Yes, let’s go with that 😉

@gkay92 @GISAID @TAlexPerkins Thanks Karthik! It was all made possible with your excellent mentorship 😉

This was a big team effort only made possible by global 🗺️ sequencing efforts and the team at @GISAID! In particular, I wanted to highlight @malkuzweny's work in building the R package. She has gone on to doing more exciting work as a PhD student in @TAlexPerkins's group!

We've also described the technical details on outbreak.info in these preprints, Genomic reports medrxiv.org/content/10.110… Research Library biorxiv.org/content/10.110…