social.coop is one of the many independent Mastodon servers you can use to participate in the fediverse.
A Fediverse instance for people interested in cooperative and collective projects. If you are interested in joining our community, please apply at https://join.social.coop/registration-form.html.

Administered by:

Server stats:

492
active users

#bioinformatics

19 posts18 participants6 posts today
Replied in thread

@thatdnaguy
"As a field, we’ve incentivized poor proxy metrics instead of research rigor. The increasing number of published papers and retractions makes it crucial to support and incentivize research rigor instead. One part of rigor is maintaining original, unaltered data that underlie key results. Researchers often don’t understand what constitutes original data or how to preserve it for future use. This job may then fall on collaborators with more expertise in informatics."
#bioinformatics #OMICs #FAIR #ReproducibleScience

Next-next experiment might involve using adaptive sampling on the ONT platform. I'm curious, did anyone try adaptive sampling with older mid-GPU models, like 3060? Wondering if slower real-time basecalling on those cards will impact the output in someway.

I'm thinking @Psy_Fer_ might have the answer to this!

🟣HealthDataScience Day 2025 is coming!🟣

Mark your calendar for 6 November & join a free #conference on all things #HealthDataScience!🚀 Hosted by #UCPH #HeaDS, this event is packed with exciting #speakers & #activities!
We will share details bit by bit - there is too much exciting information about speakers and activities to share all at once!

🗓️ 6 Nov 2025
⏰ Reg opens 1 Apr
🌍 heads.ku.dk/calendar/2025/even

Is there a linux command-line tool to copy files while specifying the source and *target path of each file*?

'rsync' can read an input list of files, but has no way of specifying individual target paths.

I am thinking of using the '--transform' option of `tar` and do a tar - untar pipe, but it would only work for short substitution scripts and when you can figure out the target given the source path only.

I've been reviewing state of ONT & PacBio #Archaea raw data availability in public repos (preprint coming soon). It's not fantastic.

Once we upload current reads, our warehouse & partner lab might be THE top uploader in the world in terms of diversity of Archaeal species contribution.

We're also covering both #ONT and #PacBio for each species - so 2x the count.

I'll also upload all ONT raws in blow5 format as well! Down with vendor lock-in.

As I face the prospect of analyzing a dataset with 3 variables: time (3 time points), temperature (2), exposure to chemical (2); but only 2 biological replicates, please, people, consult a biostatistician before doing your experiments.

Please, please, please, talk to someone in data analysis *before* running these kinds of experiments. If there isn't enough money for *more samples*, drop one of the factors so there are more "N" for the remaining ones.