*Thread Reply:* Commentable atm. I was a bit wary with sharing an editable version to all but feel free to request edit access!!
Ahaaaa I found what I had mentioned at iSBA: https://microbe.land/97-metagenomics-classifiers/
But that was from 2018 by that looks of it!
*Thread Reply:* Haha fair enough!
*Thread Reply:* And maybe leave a comment after the one on #general that if anyone else wants to get involved they should join <#C05SQA7RA81|yet-another-taxonomic-profiler>
Btw feel free to rename the channel/project! 😅
*Thread Reply:* I was working on the fly
*Thread Reply:* No worries! This all is great! Thanks for all the help!! 🙇:skintone4:
*Thread Reply:* It is very likely that I had that in my mind actually, I litrelly 5 minutes ago opened that to add a comaprison to the taxprofiler preprint 😆
I forced myself to go on the birdsite, just for you @Aleksandra Laura Pach 😉
*Thread Reply:* Haha, sorry! I should have said earlier... But i had also asked if anyone knew what I was looking before, and I realised it would be mean not to post it if I found it ..
But at least you can delete the account now😂
*Thread Reply:* Hahah, no worries. You are absolutely right, and we got an answer out of it 🙌
And I might, some wise men say it's better to have one.. I'm still very hesitant 😬
*Thread Reply:* I would've said 6 months ago that yes it's better to have one, it's very good for networking. But now it's a pile of poo.
Mastodon is much better, the way it works out keeps you on one topic so it's less 'addictive' and currently a lot of scientists are there
*Thread Reply:* Alright alright!! Good intel! Thanks
*Thread Reply:* I'll try to be there too! But it's a bit of a funny time for me as I'm in NZ, so don't wait for me if I'm not there haha.
*Thread Reply:* ❤️ you really Maria
*Thread Reply:* (but I will of course accept 🍪 s from both of you next time I come through Paris/Copenhagen to compensate my pain 😏 )
*Thread Reply:* Are :doggo: hugs also allowed as compensation?
*Thread Reply:* I'll allow it
Also in a separate comment the same guy asked in the context of nf-core/taxprofiler about custom database building... https://genomic.social/@PhilippBayer/111140387294382855
Once the idea of this project has been formalized maybe we could borrow the idea and record something similar?
*Thread Reply:* Abso-heckin-lutely! The DB creation is tightly linked to the overall performance and user experience of the profiler so we should definitely look into this!
*Thread Reply:* I think simply 'can you build custom databases' would be a big boon
https://app.gather.town/app/PlXjb0deog0B4JCq/spaam-community
Meeting notes if anyone wants to help: https://docs.google.com/document/d/1FfRCjMYO-HHbbqqKsHZipsNYjPvhzF9mIgEas8yJrUw/edit?usp=sharing
We tried a few years ago to add a aDNA challenge dataset to LEMMI benchmark, but that never really took of.
https://gitlab.com/ezlab/lemmi/-/issues/5
*Thread Reply:* This is a good point! I think it is more of a pipeline tab than classifier one if I understand correctly? But I could definitely add that in in one way or another!
*Thread Reply:* Yes indeed I think it would fall under the pipeline category!
*Thread Reply:* I'm not sure I follow... Do you mean just whether the pipeline does the trimming/merging for you?
*Thread Reply:* That and how it was done too bc it may affect how the classification it’s done
*Thread Reply:* Yeah reasonable!
Does anyone know of standalone tools that simulate deamination on the ends of reads?
We may already have a route into a benchmarking classifiers on aDNA 👀
https://gitlab.com/ezlab/lemmi/-/issues/5#note_1591209730
*Thread Reply:* You can run it as a separate command?
*Thread Reply:* Yes! I guess not for fastq but fastq but one can get around that. Here is in general how I did it for my thesis simulation: https://github.com/AleksandraLaura/thesis/blob/main/1.%20Data%20prep/1.1.%20Simulation/commands.txt|https://github.com/AleksandraLaura/thesis/blob/main/1.%20Data%20prep/1.1.%20Simulation/commands.txt
Looks like thanks to @Maxime Borry and you guys, we are getting aDNA benchmarking data in LEMMI!
https://gitlab.com/ezlab/lemmi/-/issues/5
*Thread Reply:* That's amazing! Thank you for taking on the correspondence and explaining things so well in there! 🙌
*Thread Reply:* Again, something I can do single handed 🤣
*Thread Reply:* But @Maxime Borry please chime in if I've said anything wrong/you have better ideas (or anyone else for that matter)
*Thread Reply:* Haha no, let's see how it takes to have it implemented by the LEMMI people 😉 As you saw, my original issue is already almost 4 years old
*Thread Reply:* A couple urgent project took over lately to be honest, but it’s not a standstill - just slower progress!! The sheet is almost ready and I’m figuring out whether/how to move it over to GitHub!
*Thread Reply:* would it be possible to meet after 15 CET? I'd really like to join, but am in California. Completely understand though if that doesn't work for most people/if I'm the only one in this situation
*Thread Reply:* It would be perfectly fine for me! Let’s see which day gets the most votes and see what everyone says. But please just vote regardless of time! I’m fine with meeting at 10pm CET if it means that a lot of people can join 😅
*Thread Reply:* I will likely have to leave early for Kita pickup but otherwise 👍
Could you make a google calendar invite 🙂: jfy133@gmail.com
Hi everyone! The day has finally come and I'm very exited to meetup at 15 CET and talk about what has been done so far and even more importantly talk about what needs to be done. We meet in Gather (the link below). 🥳 https://app.gather.town/app/PlXjb0deog0B4JCq/spaam-community
*Thread Reply:* Due to overlap ping appointments I might only be able to attend the first 30mish :(
*Thread Reply:* That's alright. You can overload us with input for the first 0.5h and then we can process it for the rest of the time 😁
*Thread Reply:* Will let you know if I can't join at all
*Thread Reply:* I will be travelling back on the 8th maybe it’s worth making a date survey after the day survey 😉
*Thread Reply:* I can possibly make any but I have no idea where I will be on each of the days so I don't want to influence the result
deamination patterns has been added to the new upcoming LEMMI classifier benchmarking tool (V2)!!
https://gitlab.com/ezlab/lemmi/-/issues/5
Fragmentation isn't there yet, but would someone be interesting/willing to test it out to see how 'realistic' the simulatd ancient mode looks like?
*Thread Reply:* @Maxime Borry you might be interested in particular as the original author of the idea / that thread 🙂 )
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10782538/
*Thread Reply:* Here is 152 for you @Aleksandra Laura Pach 😉 https://joss.theoj.org/papers/10.21105/joss.04360
Hey @channel! 👋 This is just a reminder that tomorrow I’m going to be available on gather for the whole day to work on filling out our sweet tool overview! Join in your own capacity but I’m super keen on getting to it with you! (If someone already knows that you will join and at which times can you please reply in this thread so I can plan my day accordingly). See ya!
}
*Thread Reply:* I'll be probably joining for about 30m some point during the morning
*Thread Reply:* I'll also join from around 30min or so at ~10am PST. No need to be there at that time though given the time difference.
Good morning! ☀️ Here is where I will camp for the day:
Oh btw, another one to add (featuring me!): https://www.biorxiv.org/content/10.1101/2023.10.20.563221v1
*Thread Reply:* I can join this time 🎉
*Thread Reply:* For a coule of hours anyway 😬
*Thread Reply:* Can only confirm on the day, depends what Aida has scehduled that week
*Thread Reply:* Unfortunately scheduling conflicts with Aida mean I can't do it on Monday. But I'll do a little bit on Tuesday too make up for it
*Thread Reply:* Absolutely completely fine 😇
*Thread Reply:* Going to connect after lunch ❤️
*Thread Reply:* Baby duty again today unfortunately :( will try to catch up a bit this week again (but getting a bit buried again)
Hope you feel better soon 🤞