Phillip Kyriakakis

12.4K posts

Phillip Kyriakakis banner
Phillip Kyriakakis

Phillip Kyriakakis

@breakliquid

Senior Research Scientist at Stanford Bioengineering | Taking apart and building biology.

Stanford, CA Katılım Mart 2009
4.9K Takip Edilen6.4K Takipçiler
Oded Rechavi
Oded Rechavi@OdedRechavi·
When are you more concerned about the future of your job, during relaxed periods at work or when you're super busy?
English
15
1
38
14.8K
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@ATinyGreenCell I find Threads to have more content I like. Not much science on any of these sites compared to the old Twitter though, so I use them all less.
English
1
0
2
104
Sebastian S. Cocioba🪄🌷
Sebastian S. Cocioba🪄🌷@ATinyGreenCell·
I'll try my best to cross post on blusky, but nowhere can I better serve small science acounts, share broadly, and also get rapid feedback on my own work than on this hellsite. 99% of my income now comes from research contracts established here so I am indebted to this community.
English
4
1
76
4.3K
Patrick Hsu
Patrick Hsu@pdhsu·
americans use mm.dd.yy europeans use dd.mm.yy scientists use yy.mm.dd
English
13
8
153
23K
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@ATinyGreenCell Addgene usually excludes the gene of interest from the feature list/map. So annoying for users, super bad training data for AI.
English
0
0
1
169
Sebastian S. Cocioba🪄🌷
Sebastian S. Cocioba🪄🌷@ATinyGreenCell·
I'm sorry but I have a hard time accepting this as a worthy advancement. Training on addgene plasmids, which many are running legacy vectors with a LOT of excess junk from back when we didn't have a choice is gonna inspire a lot of the same dated design conventions. Hard disagree
Biology+AI Daily@BiologyAIDaily

PlasmidGPT: a generative framework for plasmid design and annotation 1/ PlasmidGPT introduces a novel framework for designing plasmid DNA using a transformer-based generative model. It leverages 153,000 plasmid sequences to create new designs with low sequence similarity to the training data but with similar structural properties. 2/ One of the most exciting features is PlasmidGPT's ability to generate plasmid sequences based on specific user inputs, enabling controlled and custom designs. This opens up new possibilities for efficient plasmid creation in biotechnology applications. 3/ The model uses sequence embeddings to predict attributes like lab origin, species, and vector type, outperforming conventional models like CNNs. It demonstrated an 81% top 1 accuracy in predicting the lab of origin, a significant improvement over previous approaches. 4/ PlasmidGPT also successfully extends its capabilities to natural plasmids, showing comparable performance to CNNs, but with much faster computational time. This could facilitate rapid analysis of both engineered and natural plasmid sequences. 5/ A fascinating aspect is its application in predicting unseen plasmids. In out-of-distribution tests, PlasmidGPT accurately predicted the lab of origin and growth strain, demonstrating robustness in handling new data. 6/ By treating DNA sequences like language, PlasmidGPT provides an intuitive, efficient tool for plasmid analysis and design. Its ability to fine-tune for specific applications (e.g., mammalian vs. bacterial plasmids) highlights its versatility. @ShaoBin_phy 💻Code: github.com/lingxusb/Plasm… 📜Paper: biorxiv.org/content/10.110…

English
9
23
197
29.1K
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@HarmitMalik Man, i feel that way with so many fewer requests than you. I can’t imagine how you do it.
English
0
0
1
266
Harmit S. Malik
Harmit S. Malik@HarmitMalik·
It is academic letter request season.
GIF
English
5
6
73
8.2K
Sebastian S. Cocioba🪄🌷
Sebastian S. Cocioba🪄🌷@ATinyGreenCell·
Honestly, I don't blame anyone, neither my Q5 or the nanopore flow cell pores for not figuring out wtf to do with this polyA stretch, lol. I'll keep the clone that reads closest to truth, knowing it's either a sequencing artifact or my Q5 said "this is enough A's for one run"...
Sebastian S. Cocioba🪄🌷 tweet media
English
10
1
32
4.5K
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@ATinyGreenCell I can also use my fluorometer! I can pick the bandwidth and wavelength. It is super easy to stick things in there!
English
0
0
2
111
Sebastian S. Cocioba🪄🌷
Sebastian S. Cocioba🪄🌷@ATinyGreenCell·
A dear friend donated an old plate reader to the sucrose project and now spending my lunch break inspecting parts and seeing what software is needed. This will act as a good benchmark for my bespoke plate reader build next month. Powers on but screen is busted. Checkin ebay...
Sebastian S. Cocioba🪄🌷 tweet mediaSebastian S. Cocioba🪄🌷 tweet mediaSebastian S. Cocioba🪄🌷 tweet media
English
4
0
20
2.6K
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@bielleogy @ATinyGreenCell I do him most start off with some integrity and most bad actors learn to be bad. So there are many examples you hear about of people that go from bad to good. That may also be bc of negative bias in our stories and memories too.
English
0
0
0
51
Sebastian S. Cocioba🪄🌷
Sebastian S. Cocioba🪄🌷@ATinyGreenCell·
There is no grand genius when it comes to biology. No predisposition for bioresearch greatness. No BioMozarts. No BioJordans. The challenges are too massive and any semblance of extreme competence comes from very hard work and a LOT of luck. Natural talent for bio is myth.
GIF
English
17
13
156
12K
Proteios Technology, Inc.
Proteios Technology, Inc.@ProteiosTech·
@breakliquid Great question Phillip! We use L-Lysine and NOT imidazole. We're having some grand opening promotions today, if you'd like to try them out for yourself!
English
1
0
1
49
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@com_ecology @tomkXY One crazy part is that senior PIs have so much less to gain by a Nature paper (or another Nature paper!), but they keep up the system and expectations. Jr researchers are dependent on them for promotions in most places, or to even get funding and keep their jobs.
English
0
0
1
22
@comecology.bsky.social
@comecology.bsky.social@com_ecology·
@breakliquid @tomkXY Agreed! not simple. Some journals "justify" their exorbitant APCs by saying - hey, it's s an investment in your career bc our journals "assure" jobs, promotion, recognition & grants. Those paying "bet" in such strategy & convince themselves in some ways that's ok! Troublesome!
English
1
0
1
31
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@haesleinhuepf Its a big part of the answer, but more specifically senior PIs need to not submit to Nature et al, and keep the manuscript as a preprint first. They are the ones that can usually afford these fees and they set the example.
English
0
0
1
103
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@tomkXY If only it was that simple. It takes many of the senior PIs to do that first and normalize it before others can do that and still get promoted.
English
1
0
1
92
Phillip Kyriakakis
Phillip Kyriakakis@breakliquid·
@ambulanzen @LingfengGe “because our industry is, to a large extent, resilient to economic downturns as investment in research and learning continues.” Actually because its a monopoly, its resistant.
English
0
0
1
17