BasesAndBits

@d3mOcr4cy_exit @curiouswavefn Any luck identifying folks in this space? I'm also interested in the space and have some initial thoughts

@curiouswavefn Who is thinking about how we incorporate AI into STEM graduate education? We need to develop a pathway to train the brains of the next generation or we risk ceding our scientific autonomy to machines. I’d love to follow folks working on this.

I've always found it odd when people say that "AI will do all the tedious stuff while scientists and researchers do the advanced stuff." The problem is, you can't really reach the level of doing the advanced stuff without training yourself to do the tedious stuff. My generation might be the last one who learnt to do the tedious stuff without AI. It's going to be really interesting to see how the next generation handles this challenge.

@simocristea @NCIEytanRuppin This doesn't look reasonable at all. Maybe paired with actual expression data or other functional data it'd be possible. But to make any decisions with such low correlation would be unreasonable, no?

Few people are aware that expensive Spatial Transcriptomics gene expression data can be reasonably predicted from the super cheap H&E slides. e.g. Path2Space @NCIEytanRuppin

@gama_search @AdrianoAguzzi Out of curiosity, what tasks using bedtools and VCF tools do you want gpt to do?

@AdrianoAguzzi Or a dedicated agent who can automatically use deseq / bedtools / vcftools and do correlation studies ;)

@fchollet This isn't anything novel, it's just adding feedback to the prompt based on what the previous output looked like.

Jeremy Berman, former #1 on the ARC-AGI-Pub leaderboard, just open-sourced his solution as a template on Params (link in next tweet) It uses LLM-driven genetic programming, which has turned out to be way more powerful than anybody expected. You can book a consulting call with Jeremy via Params to talk about how the same approach could apply to your use case.

@JoshRainerGold How do you have crippling back pain and be so swole? Maybe those are not recent pics, but the situation devolved so quickly (within some yrs) that he did something so drastic?

All discussions of motive for Luigi Mangione have to start with his back pain. Anyone who has lived with it can tell you how debilitating it can be. There’s no relief. You feel it constantly, such that your back starts to numb to ease the pain. You’re prevented from not just enjoying your favorite activities, but living life in the moment. You’re always waiting for it to hit again, and that’s if it ever lets up at all. It can seriously impact your sleep as well. All my friends that had back surgeries say it never helped, but it’s the most common prescription. This is a young fit man who loved to hike and surf. That likely was taken from him and he sought relief any way he could. Everyone I know from the gym has looked into Dr. Stuart McGill and his “Big 3.”

@R_Graph_Gallery There's nothing wrong with the histogram! You can't choose a correct bin size. You can change the bin size to generate more than one histogram though. That's easy. Reading the histograms is also easy.

Friday afternoon quiz 😊 ➡️ What's completely wrong with this histogram? The graph shows the distribution of marathon finish times for approximately 400,000 men. I'll add a before / after in the comment in 1h 🙃

@RadioGenoa Opportunity, gain wealth, be free, have family, interested in higher education...

There are 57 Islamic nations in the world, why do Muslims come to Europe and America?

@yaroslavvb @doodlestein Complex -> rotation

@doodlestein That intuition works well for normal matrices. Works slightly less for non-normal matrices with real eigenvalues. Meanwhile a large random matrix has complex valued eigenvalues. If the eigenvalue is "i", is the space "stretched" or "squished" in that direction?

Where do eigenvalues come from? They don't have a nice geometric interpretation. For triangular matrices, eigenvalues are just the diagonal entries. Off-diagonal part of the matrix is ignored even though it has an influence on geometry 1/6

@vsbuffalo Dude idk you but you're ridiculously proficient and productive. Idk how you do it but man I want to be able to solve problems and make it appear easy as you do

Today's project: find the factor of two error in ~4k lines of Rust.

@byu_sam Allen brain institute has large datasets of brain images

What large health care datasets exist? I know about AllOfUs, Biobank and FinnGen. Are there others like this? Are they all a mix of medical/molecular data?

@Pinolero1986 I'm guessing you were on vacation? I feel like stress associated with actual life is at play here

@TAYVAY_ This is the whole point buddy. We don't want unchecked investing in all the property by those who have capital. We want everyone to be able to afford housing near where they have family and their job. Work at target? Best believe you should be able to live near it w/out struggle

This is why investors are scared of investing in California: City of Santa Monica just passed a law limiting annual rent increases to $76. On any rent...in any unit...of any quality. The best example I had to laugh at: Tenants in the 16-story Champagne Towers on Ocean Avenue will see a $76 increase. The highest monthly rent was $41,812 and was paid by a tenant who has occupied a three-bedroom unit since 2016. That tenant will see an increase of 0.0018 under the $76 cap. Makes you wonder when it will become uneconomical to own a building in Santa Monica. For many, probably already is.

@HarryHu5268 @sitzikbs @CVPR I'm sure this "political" issue can be proven as genocide

@sitzikbs @CVPR You are absolutely right. Science is something that can be proven or disproven, but politics is not.

I was deeply offended by a slide in a recent talk at #CVPR2024 that falsely accused my country of genocide. Such baseless political statements have no place in our scientific community. Let's keep our focus on advancing science and leave politics at the door. @CVPR

@sitzikbs @CVPR Israel is quite literally conducting genocide

@jchal Which has the best weather?

Five metros that will explode in the next decade: 1. Savannah, GA 2. Wilmington, NC 3. Chattanooga, TN 4. Myrtle Beach, SC 5. Panama City, FL What did I miss?

@johnmark_taylor Is this a product of e's Taylor expansion??

Some magic that I overlooked till now: raise e to the power of the negative squared distance between two vectors. This ends up being equal to the dot product similarity between all possible interaction effects among the vector entries (w/ higher-order interactions weighted less).

@cremieuxrecueil p-values adjusted for multiple testing?

What's more convincing? p = 0.04 in a sample of 10 or p = 0.04 in a sample of 1,000,000? 🧵

@Caroline_Bartma My first time using a nanodrop I legit thought you have to drop a nanoliter in and I was scratching my head like how tf can that be done. All labs should have some better training than what I went through lol

@KonstantinKisin @RIKKISCHLOTT Lol it's not hard to read. That's like saying you know global warming isn't happening but I didn't read about it because the effort was futile. Read it for yourself then make a decision, no?

@RIKKISCHLOTT I tried to read it and gave up after an admirable but ultimately futile effort.

Huberman literally just ignored this NY Mag story and successfully let it whither. Amazing.

I'm asking those who love the outdoors to post a picture YOU took. Just a pic. No description. The goal is to regain peace and harmony without negativity. Please copy the text, put a picture on YOUR page, and let's look at these beautiful pictures. 💙💙💙