@jpelbers @jeroen_deridder @nanopore @UMCU_CMM @oncodeinstitute If you know where exactly the modification is, and the change in signal is sufficient, it should be applicable to other modifications.
English
Marc Pagès-Gallego
779 posts
Marc Pagès-Gallego
@marcpges
Utrecht, The Netherlands Katılım Mayıs 2010
161 Takip Edilen99 Takipçiler
@jeroen_deridder @nanopore @marcpges @UMCU_CMM @oncodeinstitute Same idea would work with deaza-GTP ? Can you use a PCR product with 30% deaza-GTP added (or does it need to be 100%), sequenced by MinION instead of a synthetic oligo? We also have native, adaptive sampling on the same region of PCR amplicon.
English
@KubaSawicki_UWM @jeroen_deridder @nanopore @UMCU_CMM @oncodeinstitute The model was trained on R9.4.1, so it will fail on R10. Unfortunately, an R10 model would require significant changes in data design due to their larger nucleotide context, so it is not in our goals to make such model.
English
@jeroen_deridder @nanopore @marcpges @UMCU_CMM @oncodeinstitute Will this model work with R10.4.1 pores? Or any update comming soon?
English
English
English
Marc Pagès-Gallego retweetledi
Nanopore sequencing from formalin-fixed paraffin-embedded specimens for copy-number profiling and methylation-based CNS tumor classification #neuropath
link.springer.com/article/10.100…
English
Marc Pagès-Gallego retweetledi
Our R package for visualizing chromosomes, genes, SVs and copy number is published now. Usage is simple and flexible! academic.oup.com/bioinformatics… github.com/cortes-ciriano…
antisense.@razoralign
ReConPlot: an R package for the visualization and interpretation of genomic rearrangements academic.oup.com/bioinformatics…
English
Marc Pagès-Gallego retweetledi
👏Enhorabona a les estudiants de Comunicació #URV Neus Jover i Clara Pagès, guanyadores d'un dels premis Ramon Barnils pel reportatge radiofònic "Tarragona, com respires?", que qüestiona diferents aspectes relacionats amb la petroquímica de Tarragona.
👉diaridigital.urv.cat/estudiants-com…
Català
Marc Pagès-Gallego retweetledi
🚀 We're Hiring at Cyclomics! 🧬
Are you passionate about software development and data analysis? Do you thrive in a cutting-edge bioinformatics environment? If yes, then we have the perfect role for you!
linkedin.com/jobs/view/3767…
English
Marc Pagès-Gallego retweetledi
Marc Pagès-Gallego retweetledi
Un sistema d'IA anomenat Sturgeon ha estat capaç de classificar correctament 18 de 25 tumors cerebrals en menys de 45 minuts, cosa que permet als cirurgians fer servir la informació durant l'operació
nature.com/articles/d4158…
Català
Marc Pagès-Gallego retweetledi
AI rapidly diagnoses brain tumours during surgery go.nature.com/3rW8G1n
English
Marc Pagès-Gallego retweetledi
Research in @Nature! Researchers from @UMCUtrecht and researchers, pathologists and neurosurgeons from the @prinsesmaximac and @amsterdamumc collaborated. In this video, Oncode Investigator @jeroen_deridder explains the most important conclusions.
youtu.be/1Ni2oaYyPZc
YouTube
English
Marc Pagès-Gallego retweetledi
🆕@Nature
Deep learning #AI during surgery, using low cost, intraoperative nanopore sequence, to rapidly classify tumors in <90 minutes
nature.com/articles/s4158… @jeroen_deridder
English
Marc Pagès-Gallego retweetledi
New #NRMCBReview by @MariaRodrColman, Tobias Dansen & Boudewijn Burgering @UMCUtrecht discussing the ⚔️double-edged role of #FOXO transcription factors in stress responses
go.nature.com/3PMFbIo
English
Marc Pagès-Gallego retweetledi
The fam and I have moved back to the Netherlands 🤷🚲🌦️.
I'm fortunate to be doing a bit of work with @fulcrumgenomics and am looking for other contracting work especially for next couple of months.
Please get in touch if you have something I can help with.
English
Marc Pagès-Gallego retweetledi
Major, fatal errors found in the data and methods of a 2020 paper in @Nature, including millions of reads mis-identified as bacteria. The "cancer microbiome" in this study was simply not there. @abrahamgihawi @elapertea @YuchenGe1 @JenniferLu717 biorxiv.org/content/10.110…
English
@NatureBiotech The reference links on the third paragraph are off by one.
English
Single-cell data from RNA-seq, chromatin accessibility, DNA methylation and other modalities can be readily integrated using two new methods #NBTNV go.nature.com/4222tNu
English
@aldertzomer @jeroen_deridder @UMCU_CMM @oncodeinstitute @nanopore Yes, Bonito is the current architecture in guppy. If I am correct (might be wrong), all current guppy models use the Bonito architecture, but the super acc model is larger (and more compute expensive) than the hac and fast models.
English
@jeroen_deridder @marcpges @UMCU_CMM @oncodeinstitute @nanopore Nice! Bonito would be similar to the super accurate basecaller in guppy right?
English
Marc Pagès-Gallego retweetledi
Ultra-fast deep-learned pediatric CNS tumor classification during surgery medrxiv.org/cgi/content/sh… #medRxiv
English
@JosephGuhlin @vellamike @Hasindu2008 You could prepare the data with bonito, and add the noise in the hdf5 file later to be used by bonito.
English
@vellamike @Hasindu2008 Yeah. I do want then noise to be something the machine might generate though, so not totally random. But that's still a really good idea. I've got no experience with the signal files yet. Do they open easy with numpy?
English