Lee Organick
215 posts
Lee Organick
@LOrganick
Computer scientist, biologist, thru-hiker among other things.
Seattle, WA Katılım Mart 2017
107 Takip Edilen225 Takipçiler
New PhD thesis just publicly dropped (see if you can spot the pun).
If you've always wondered about DNA data storage but don't know much about it, pages 1-4 might be helpful! digital.lib.washington.edu/researchworks/…
English
If you’re an experienced wet lab scientist, this may be of interest!
Cache@cachedna
Exciting News: Cache is hiring a Development Scientist! 🧬 We are seeking a talented and driven translational R&D member to join our small but growing team. Apply here: lnkd.in/eu3aK4D4 #hiring #innovation #research #development #scientist #laboratory #biotechnology
English
@dinazielinski Huh, I wouldn’t have guessed there would be quantum questions! Sounds like a really great time, I hope someday I get to go to one of your talks!
English
@LOrganick Mostly about the key challenges and I told them how scientists like you are making rapid advances to address them! Also a question about competing with quantum computing
Paris, France 🇫🇷 English
Many thanks to the Royal Dutch Archivist Association for the invitation to speak about a role for DNA storage in the “Memory of Tomorrow”.
Great energy and questions!
Sylvia Bakker-Kempen@SylviaBakkerK
Great talk @dinazielinski ! ‘Nature’s oldest storage device: DNA’ #archiefdagen #archiefdagen2023
Rotterdam, The Netherlands 🇳🇱 English
@vitaliikl Basic stats like linear/multiple regression (which…I suppose is ML lite 😆).
Yep copy number= your best estimate of RNA count (really tedious, very tricky with RNA)
Yep DNA comp= GC content/length/any chemical modifications, etc.
I like your problem, I’ll keep thinking on it.
English
@LOrganick Would be good to know what you mean by basic stats, copy number (true unknown RNA count?) & DNA composition (GC content?).
I am looking for a solution that can fit the approach of adding a layer representing PCR bias without using gene-specific parameters
twitter.com/vitaliikl/stat…
Vitalii Kleshchevnikov, PhD@vitaliikl
Would be great to have a model of PCR bias that can be used as a layer in any scRNA model (think latent space estimate, NMF, sequence-based regulatory models): give it DNA sequence + biological signals estimate and it outputs NB expected value estimates after applying PCR bias.1/
English
Recent paper on PCR bias in scRNA got me thinking. Has anyone tried a sequence based deep learning model of PCR amplification biases (incl how the number of cycles affects the bias)? Intuitively, modeling this seems possible - PCR is a fairly standardised&simple biochem reaction.
English
@vitaliikl Context if helpful- PCR bias depends on many known variables, eg. copy number, cycles used, DNA composition
If you know all that, then using basic stats is gonna get you there just fine & give a more interpretable model too. If you don't, well, garbage in, garbage out.
English
@vitaliikl Just that I agree with your initial tweet that "modeling this seems possible - PCR is a fairly standardised&simple biochem reaction".
Trying to use deep learning for PCR bias is like using a jumbo jet to go to the grocery store. Really overkill and ultimately not helpful ;)
English
Lee Organick retweetledi
Hi friends, I've started a crowdsourced Google Doc for open positions and helpful volunteer contacts to help laid-off Googlers find jobs. Please add any openings you know of, and list your contact information if you're willing to help!
docs.google.com/document/d/1g5…
Boston, MA 🇺🇸 English
My first journal cover graphic is out, spotlighting our #DNA #nanotechnology based multiplexed #biomarker detection system! Made with @UCSFChimeraX (protein + DNA), @autodesk Maya (3D scene), and @affinitybyserif Designer (finishing)! 🎨 + 🧬
Analysis & Sensing@AnalysisSensing
#OnTheCover Multiplexed Label-Free Biomarker Detection by Targeted Disassembly of Variable-Length DNA Payload Chains (Dunn) onlinelibrary.wiley.com/doi/10.1002/an… @kdunnresearch @matt_aqui @EdinburghUni
English
Lee Organick retweetledi
I am so excited to announce the 2022 @bitsinbio x Cradle Bio Biodeveloper Survey!!!!
This survey is inspired by the Stack Overflow dev survey and has 3 purposes:
- community
- tools
- problems
Everyone who works in life sciences or software for life sciences should participate!
English
@LOrganick Would love to connect about the Chief of Staff role!
English
Our company is #hiring! We’re looking for a backend-engineer, research associate, and chief of staff. The team is great, the project is fascinating, please share with anyone you think might be interested! jobs.lever.co/cache-dna/
GIF
English
Lee Organick retweetledi
Adaptive Biotech posted internships earlier than usual this year! (We're trying to switch to the tech intern schedule.)
It's a great place to work with lots of interesting problems (especially in Antigen Map, totally not biased)! Hybrid or fully remote: adaptivebiotech.com/career-listing…
English
Lee Organick retweetledi
If the US wants to maintain deep tech leadership, they should ease the burden of immigration or risk losing talent to competition. My brief perspective: sciencepolicyreview.org/2022/07/mitspr…
Thanks to @SciPolRev @Knappe_ @BertNeyhouse
English
Can’t believe my time at @datamolecules is coming to an end, I couldn’t have made it without the patience and help of every single person there, not to mention my amazingly supportive cohort. *Thank you *! @luisceze @KStraussIdeas @jeffnivala thank you x ∞
Luis Ceze@luisceze
Congratulations Dr. @LOrganick! You ventured into a PhD between CS and molecular biology, done seminal work in DNA data storage, and were instrumental in getting UW MISL @datamolecules off the ground! So *Thank you* and Go Lee! H/T to co-advisors @KStraussIdeas and @jeffnivala!
English
@Georgeos_ And lastly there’s this amazingly cuddly cat who we met on a walk once and all she wanted to do was lay on my lap
English
