Can scientists be taught how to choose good projects?
I took a fantastic class taught by
@mfgrp
last quarter where the foundational hypothesis is: Yes, they can!
And I continually come back to these core insights as I think about project choice and direction in my 2nd year of…
I’ve been super into the idea of intracellular molecular recording to measure cell dynamics, but this tech offers an interesting alternative: make the cells give you a sample!
Discovery of a new class of antibiotic using AI, shown to be effective against MRSA in mice!
Incredibly cool work from Jim Collins’s lab at
@MIT
in
@Nature
Calling of all 20 amino acids by nanopore! An exciting step toward nanopore protein sequencing
Beautiful work from Shuo Huang’s group at Nanjing University
Evo: a foundation model for genomics!
Using a well suited architecture, Evo learns from billions of bp of genomic sequence and performs well on several zero-shot prediction tasks on RNA, DNA and protein.
Beautiful paper from
@BrianHie
&
@pdhsu
at
@arcinstitute
A ribozyme that cuts DNA - programmably!!
It has a ways to go before it can be a precision genome editing tool - short target recognition sequence, doesn’t bind dsDNA well - but an exciting discovery nonetheless!
Study from
@GogoLiu
lab at
@Tsinghua_Uni
, published in…
Designing new DNA binders!
Very cool new method from
@cam_glasscock
, David Baker & team to computationally design sequence-specific binding proteins, some of which are demonstrated to function as repressors or activators
CAR T cells to reverse - and even prevent - aging phenotypes!
The CAR Ts target uPAR, which is upregulated in senescent cells. Treated mice see metabolic improvements & better exercise capacity, & this can work prophylactically, too: treat young mice, and the have better glucose…
Neat new system to evolve protein-protein interactions - while altering both sides of the interface! Watch for a description of the clever method, read the paper to get the deets on how these interfaces evolved & predictive models trained on screen data.
Love this idea for scaling up genetic screens (without scaling up cost!): make more perturbations per cell and recover the individual perturbation effects!
Angstrom scale resolution with a fluorescence scope?!
To think this was done on the same microscope we used for regular old imaging in my old lab…🤯
Nice work
@JungmannLab
!
Interesting approach for acne treatment: engineer bacteria to work like Accutane, but locally (and without all the nasty side effects)!
Modified C. Acnes bacteria secrete NGAL to kill sebocytes & thus sebum production. They engraft on mice skin & produce NGAL there.
Membrane editors!! Cool work in
@NatureChemistry
from
@basktastic
lab
I didn’t have time to detail their nice screening approach, or the editor’s light inducibility - worth a read!
This paper was drawn to my attention on 3 separate occasions in the last couple weeks, so ofc I had to make a video.
Incredibly cool work from
@CellRaiser_
lab in
@biorxivpreprint
!
Making blood tests for cancer detection more sensitive by making circulating tumor DNA stick around longer!
Two nice strategies detailed in this paper from
@MIT
@broadinstitute
in
@ScienceMagazine
Estimating the biological age of 11 organs via blood testing + proteomics!
Really cool study led by
@JarodRutledge1
and
@hammy_oh
from
@wysscoray
lab, with exciting avenues for future use in the clinic & research
1. The time spend selecting and planning a project is minuscule compared to the time spent doing it. It makes sense to spend a little more time planning to feel confident that your execution phase will be well spent.
Restored B & T cell populations & improved response to viral infection in aged mice via depletion of a dysregulated stem cell population
Very cool proof of concept from Irv Weissman's lab!
(s/o
@MorrealeBrian
for assisting w this video!)
I’ve been surprised by the ways in which making TikTok videos improves my own science.
The practice of creating clear narratives is useful not only for giving talks & publishing, but also for the planning and shaping of research projects.
Read more of my musings on this topic…
2. So you want to choose a good project. What is “good”?
We often equate “good” and “impactful” in science. But a potentially impactful idea is no good if it’s impossible. Designing better projects is about moving your initial idea up and to the right on this plot.
4. Importantly: Pick ONE of these project types! If you’re making a new technology, demonstrate it with the best application you can think of. If you’re answering a basic science question, be open to using any technological approach to answer it. Don’t over-fix your parameters.
Our new preprint describes “CROPseq-multi”: a versatile solution for multiplexed perturbation and decoding in pooled CRISPR screens Reagents on Addgene:
Just made my first video with an AI avatar and... wow🤯
Perhaps still in uncanny valley but there are parts where it's totally convincing! See for yourself:
Great storytelling from writers like Siddhartha Mukherjee and Atul Gawande sparked my earliest interest in science.
Hopeful that more scientists will begin to weave their experience & expertise into great stories (maybe with video, too?😉)
The Fall and Rise of Science Writing 📈
In the not-so-distant future, those with intense mastery over a subject will be the greatest writers.
Scientists: Now is the time to start a blog.
Thrilled to see this work out in the world!!
TLDR; Large scale genetic perturbations + cellular imaging helps us discover new biology.
Check out our preprint -- and yes, I'll make a video on this at some point 😉
As promised, here is a brief tweetorial for our preprint “A genome-wide atlas of human cell morphology” that I shared last week - there’s too much to cover it all here, but I’ll hit a few of the high notes
7. You aren’t done once you’ve picked a project.
Things fail, or stall, or are more challenging than anticipated. Sometimes, it’s critical that you lift your head up from your lab work and ask yourself: Am I still on the best path to answering my question? Or sometimes: am I…
Are any undergrads/other students interested in working with me on videos? Specifically on video editing, image generation, paper sourcing
My DMs are open!
5. Ok, you now have an impactful new technology or biological question to work on. But if it’s really impactful, somebody has probably worked in this space before. What’s novel about your approach?
Great projects in biology often have some core kernel of novelty: it may be…
Really cool to see Gordian’s platform launch today. I always thought in vivo perturb-seq seemed nearly impossibly hard to use in a drug development setting, but they’re doing it! Finding targets!
Excited to see where it goes!
Shouldn’t aging be part of drug discovery for diseases of aging?
Gordian exists to move therapeutic testing from plastic wells and young mice into a living environment with all the complexity present in patients with age-related disease.
We’re now…
Kidneys from genetically engineered mini pigs were transplanted into monkeys, and some persisted for years!
An exciting update from
@eGenesisBio
in
@Nature
this week.
6. Alright, you have a great approach to solve an interesting problem - but how many other people may have that same idea? It’s good if there’s potentially a little competition (“fish where the fish are”), but not good if there’s serious competition and you aren’t well equipped…
In case you missed my first solicitation -
If any students would like to work with me on vidoes (scripting, paper sourcing, making images), shoot me a message!
Good writing is good insight, packaged clearly.
It's a lesson I must always remind myself of when writing, and one nicely hit upon in Niko's advice for blogging.
Highly recommend a read!
Today marks 3 years since I started my blog.
Writing on the Internet helped me form my deepest friendships. It helped me get several jobs.
And I hope you will consider writing, too. Here's my advice on how to write a blog & keep it going.
*****
I…
If you like my videos, I request your input! Help me do some market research & let me know…
- What’s your professional background?
- What value do you derive from my videos?
- Are you more or less inclined to read the paper after watching?
Feel free to respond here or DM!🙏
Asimov Press, a new publishing venture that features writing about biology, launched today.
Our magazine and books will cover everything from biosecurity to vaccine development & the long arc of progress in genetic engineering.
Learn more & subscribe:
Our first limited batch of Firefly Petunias is set to bloom in Spring 2024 — and we cannot be more excited to announce the start of orders in the US!
Order your Firefly Petunia at
Been looking forward to this for months!
@mfgrp
summarizes the lessons taught in his excellent class on problem choice in science & eng. Highly recommend a thorough read.
✅ HammingBio Day 3
It’s a wrap! Thank you for putting so much energy and love into these last sessions. This HammingBio unconference won’t be our last! ❤️
Thank you again to our sponsors
@LongJourneyVC
@bismarckanlys
@athena_DAO_
for making this event possible 🙏
The SaaS startup playbook doesn’t work in deep tech. “Move fast and break things” doesn’t apply when building nuclear reactors or bioengineering cell therapies.
That’s why we built 5050, “a cheat code for starting a deep tech startup.” (quote from an alumni)
Applications open!…
1/ Solutions to the climate crisis are already here. But they're stuck in academic labs.
Often, after the basic science is done, there’s just a little more work needed to show the science has commercial potential. But no one funds this!
Introducing Manifest Climate!
Since I only started posting videos on Twitter in Jan 2023, I have dozens of old paper explainer videos on my TikTok that surely most of you have never seen.
Should I post them all here as well?
Lately, I've been thinking a lot about scientific communication with the public in the age of disinformation. A summary of my thoughts on how to reach more people with science content: