Genomic Technologies Group retweetledi
Genomic Technologies Group
288 posts
Genomic Technologies Group
@GenTechGp
Genomic Technologies Group // The Garvan Institute of Medical Research // Australia's leading long-read sequencing service // Enquiries: [email protected]
Katılım Ağustos 2020
266 Takip Edilen664 Takipçiler
Genomic Technologies Group retweetledi
Cornetto v0.2.0 is now released for using programmable selective nanopore sequencing for genome assembly.
GitHub: github.com/hasindu2008/co…
Paper: nature.com/articles/s4146…
Datasets: hasindu2008.github.io/cornetto/docs/…
... and a banner made by Ira @GenTechGp referring to ‘no boring bits’.
Hasindu Gamaarachchi@Hasindu2008
Our cornetto work is now published at nature.com/articles/s4146… Cornetto can do near-T2T assembly using @nanopore adaptive sampling (readfish by @mattloose) & hifiasm by @ChengChhy - with less 💸 - reference agnostic, so works for non-humans - not just blood, even saliva
English
Not long now until our Long-Read Research Symposium on November 13th. See below the amazing line-up of speakers and talks. It's not too late to register for this free event - click the link --> eventbrite.com.au/e/garvan-long-…
English
Genomic Technologies Group retweetledi
Join us for a day (Nov 13) of fascinating talks and discussions on the latest in long-read sequencing tech and related research in genomics, transcriptomics, and epigenetics. A great opportunity to network and hear from leading experts from around Aus.
eventbrite.com.au/e/garvan-long-…
English
Genomic Technologies Group retweetledi
Tha Garvan long read research symposium is BACK! Sponsored by @nanopore and @PacBio this event is FREE to join. Learn all about the fun you can have with long reads 🤓
register below
eventbrite.com.au/e/garvan-long-…
English
Come join us this November 13th for the Garvan Long-Read Research Symposium! You'll hear about the wonderful things you can do with @nanopore and @PacBio long reads from a great line-up of speakers. FREE to attend. Register below
eventbrite.com.au/e/garvan-long-…
English
Check out our detailed evaluation of SLOW5 vs POD5 format for raw Nanopore data. TLDR: SLOW5 still the best nanopore data format.
Hasindu Gamaarachchi@Hasindu2008
For many of those who were asking on BLOW5 vs POD5 for nanopore signal data, here is a finally detailed benchmark we did: biorxiv.org/content/10.110… Summary: performance of BLOW5 is >= POD5 (from ~= to 100X, see below), with benefit of having ~3 dependencies instead of >50.
English
Check out the excellent line-up of speakers we have for the @GarvanInstitute Long Read Research Symposium day
It is FREE to come along - this November 7th. Reserve your spot eventbrite.com.au/e/garvan-long-…
English
Not long now until the @GarvanInstitute Long Read Research Symposium on November 7th. We will have talented speakers covering the advantages of both @nanopore and @PacBio data
Only a few spots left - don't miss out! 💃
eventbrite.com.au/e/garvan-long-…
English
Cut your Nanopore raw data files in half with our new compression method ex-zd. Thanks @Hasindu2008 for leading this project 🧬
Hasindu Gamaarachchi@Hasindu2008
Introducing ex-zd, a lossless+lossy signal compression for @nanopore signal data. While lossless can only save about 1-3% over vbz, lossy can cut the file sizes by almost half with no noticeable impact on basecalling or methylation calling accuracy. biorxiv.org/cgi/content/sh…
English
@mprous1 @Hasindu2008 @nanopore That’s a good point, however we don’t think this will happen because the bits that are being removed in our 3-bit-reduction appear to encode noise, not signal (see fig 2a). We speculate that if the basecallers were retrained on bit-reduced data they may even perform better.
English
@Hasindu2008 @nanopore my concern is that with improvements in basecalling in the future there very well might be noticeable differences in accuracy between lossless and lossy compression
English
Introducing ex-zd, a lossless+lossy signal compression for @nanopore signal data. While lossless can only save about 1-3% over vbz, lossy can cut the file sizes by almost half with no noticeable impact on basecalling or methylation calling accuracy.
biorxiv.org/cgi/content/sh…
English
We are hosting a Long Read Research Symposium at @GarvanInstitute ! Our talented speakers will cover the advantages of both @nanopore and @PacBio data.
Join us to find out what long reads could do for you 🧬
eventbrite.com.au/e/garvan-long-…
English
@ClarksysCorner Two papers that sat with them for up to 6 months with nothing happening, and no responses to emails. One did end up going to review and eventually published, the other rejected. Complete shambles, and have heard others with similar stories there.
English
How long would you let a journal sit on your submitted paper without rejecting it, or sending it for review, before withdrawing it and submitting it elsewhere?
We'll hit 3 months next week.
Nb: We've contacted the journal multiple times and got replies, but no progress.
English
English
New paper out today on the cover of Science!!! Parasitic cuckoos and their hosts are locked in an evolutionary arms-race that is driving speciation. Biological collections can answer big scientific questions – in this case, how do new species form? @CSIRO @MuseumGenomics
English
Exciting news! I've taken on a new role at @nanopore as the Associate Director of the APAC Commercial Applications team. Looking forward to this new chapter and the opportunities it brings! #OxfordNanopore #NewRole
English
@drjosephpowell @jodieingles27 Definitely the money 🤑
English
@jodieingles27 That's What one of my PhD students said. So, no boat and house on the harbour, then?
English
During a PhD student coffee chat yesterday, a hypothetical came up. You can choose either a Nobel Prize for a discovery or being listed as the inventor on a patent for the discovery (which gives you $200m in royalties).
English
@Stoibs11 @hiruna72 @Hasindu2008 Hiruna can elaborate, but also pls take a look at SupNote3 for more detailed explanation of the calculate-offset feature
English
@hiruna72 @Hasindu2008 The pre-print states "we identify the base that is predicted to be most significant within a given k-mer". This seems to be the very core of the algorithm, but is not expanded upon in the text. Is this done on a per-read level? Or over many reads? Or sites?
English
Squigualiser with more features such as auto base shift enhancements and support for RNA004.
Hiruna Samarakoon@hiruna72
squigualiser has got a feature upgrade. 1. Aligning the reads to the most dominant base is now taken care of by the tool itself. Just provide --auto to your plot commands. An accurate alignment = a good coverage (>10 reads) + a long region (>200 bps)
English
@Hasindu2008 @josiegleeson @Psy_Fer_ @nanopore Hi Josie, the UHR we sequenced is actually from Agilent agilent.com/en/product/gen…
(catalog number 740000), which is from a cell line
English
@josiegleeson @Psy_Fer_ @nanopore The UHR we used is this which I think it should have mods retained
thermofisher.com/order/catalog/…
@GenTechGp could answer this better.
English
We have released some raw @nanopore data for benchmarking purposes, along with some basecalled outputs.
We use the subset of 20k or 500k for benchmarking tools and methods a LOT!
We thought they would be helpful to others too.
4/5kHz R10.4.1 & RNA004
gentechgp.github.io/gtgseq/
English
Genomic Technologies Group retweetledi
Our #slow5curl paper is out! academic.oup.com/gigascience/ar…
Big projects storing hundreds/thousands of @nanopore signal datasets on cloud storage like @awscloud #s3 will be able to save 💸,⏲️ and bandwidth while improving the accessibility of datasets to those with limited computing ..
English