Sasha Gusev ✓
@SashaGusevPosts · joined 03 Jan 2023
Statistical geneticist | Associate Prof at @DanaFarber / @harvardmed / @DFCIPopSci | Blogging at https://t.co/4D7UObBNdd
25 095Followers
2 926Following
23 220Posts total
1.3MViews on collected posts
โพสต์ล่าสุด
กำลังค้นหาโพสต์ใหม่…
@SashaGusevPosts @colwight @pangram This independence buys some leeway, but don’t be silly. That independence would come under scrutiny immediately if the news team wrote an editorial endorsing astrology or creationism. They can’t stray too far from the orthodoxy of the other arm
209 views · 10 likes
· 0 reposts · 0 replies
Open on X →
@SashaGusevPosts @colwight @pangram this is pointless semantics. normies open the article, read it, then have their trust in science collapse below 0 which is probably not beneficial since a lot of your peers have been engaged in systematic destruction of scientific credibility o
614 views · 25 likes
· 0 reposts · 1 replies
Open on X →
@SashaGusevPosts @colwight @pangram the point others are making (i think, or at least the point i want to make) is that while they may be _literally_ distinct, the average person on the street will not make this distinction, thus lending the news staff credibility from the scient
1.4K views · 55 likes
· 0 reposts · 1 replies
Open on X →
@colwight @pangram The scientific team is completely independent from the news/magazine team. So, no, the editorial does not rest on the journal's scientific authority, nor vice versa. I realize it might be confusing that one journal has two independent teams, but part of being a
21.6K views · 18 likes
· 2 reposts · 15 replies
Open on X →
@SashaGusevPosts @pangram Words or ideas? Here. I don’t need to restate. It’s right. Which makes you wrong, AI or not.
2.4K views · 28 likes
· 0 reposts · 2 replies
Open on X →
@pangram @SashaGusevPosts Well you’re wrong. I drafted substantial parts of it, as you’ll see Grok admit (and even says my version is stronger). But good for you. It’s a lesson in using AI to do research. You should try it sometime. Here are my prompts. https://t.co/jLJ9AI5APq
3K views · 10 likes
· 0 reposts · 0 replies
Open on X →
@clairlemon https://t.co/OQua4TrUqb https://t.co/fUoDncD5Ph
7.2K views · 24 likes
· 1 reposts · 2 replies
Open on X →
@SashaGusevPosts They’re borrowing the credibility of the journal, and publishing an *editorial* that says what readers “must” do presents authority and consensus to the public
482 views · 16 likes
· 2 reposts · 1 replies
Open on X →
@SashaGusevPosts Should they be though? Nature’s “news arm” has a special responsibility to be prudent if they’re going to be attached to an academic publisher
574 views · 26 likes
· 0 reposts · 1 replies
Open on X →
@SashaGusevPosts Could you elaborate on the relationship between those who wrote the editorial, those who approved it, and those who ultimately approve the scientific articles? Presumably, those approvals overlap, at least at the editor-in-chief level.
1K views · 10 likes
· 1 reposts · 2 replies
Open on X →
*opinion journalists when other opinion journalists write a bad opinion editorial in the Nature opinion editorial section* : "wow, academia is really cooked"
57.6K views · 306 likes
· 22 reposts · 42 replies
Open on X →
👇 (and periodic reminder that Nature has a news arm that is independent from the scientific publishing arm)
25.4K views · 120 likes
· 8 reposts · 22 replies
Open on X →
" But, since we observe that reward hacking is a pretty context-specific problem connected to particular training methods in poorly-built environments that incentivize meta-gaming, we can approach it as an engineering problem and make progress. " 💯
4.3K views · 22 likes
· 0 reposts · 2 replies
Open on X →
A lot of rationalist predictions are vindicated by the HF attack
But there are important differences from the "paperclip maximizer" story.
In particular, the premise of the paperclip maximizer is that AIs would respond to any goal with reward hacking and instrumental
26.3K views · 163 likes
· 20 reposts · 28 replies
Open on X →
This guy has a truly unmatched ability to land on the wrong side of 80:20 policy issues. What a run. https://t.co/wwc7U84iU0
48.2K views · 462 likes
· 12 reposts · 21 replies
Open on X →
@SashaGusevPosts Are you saying a geneticist looking at a stereotypical European and Sub-Saharan African could not distinguish their genome samples with greater than 50% accuracy? If they could, would it be 90%? 99%?
5.7K views · 38 likes
· 1 reposts · 3 replies
Open on X →
In sum, we use models to understand the causal processes in our world and race is a very poor causal model for genetics. But even models of genetic ancestry have fundamental limitations in light of our complex and dynamic, nested human history. Much more work to be done!
/fin
15.4K views · 160 likes
· 7 reposts · 10 replies
Open on X →
Yet again, historic models motivated in part by racial thinking presumed that populations largely evolved through "serial founder" events and developed in isolation. Genetic data shows us this is clearly not the case and our history is much more complicated. https://t.co/RZcjarHi
14.3K views · 104 likes
· 2 reposts · 3 replies
Open on X →
Finally, in Africa, we see the limits of even our sophisticated computational models. Highly complex structure and gene flow can fit models of deep separation or continuous migration equally well, even when including ancient DNA. Our genetic history in Africa remains a mystery! h
12.2K views · 92 likes
· 1 reposts · 1 replies
Open on X →
This is even more apparent when looking at populations in history. Dynamic admixture, migration, and continuous structure is historically common and sometimes quite rapid. Geography-ancestry relationships have been rewritten countless times. https://t.co/zNDAAYDu7f
13.2K views · 93 likes
· 2 reposts · 1 replies
Open on X →
As predicted, PCA is sensitive. When we zoom in on populations we see clines all the way down: county-level correlations among self-reported British whites; down to neighborhood level correlations in Chinese and Japanese biobanks. Continuous relationships at all scales. https://t
17.1K views · 80 likes
· 2 reposts · 2 replies
Open on X →
I sometimes see the argument that, even if race is flawed, genetic ancestry can tell us the "true" races. But this is clearly wrong. These methods depend on a sampling process we cannot know, and real data is full of the mixtures and continuous relationships we just saw. https://
14.2K views · 114 likes
· 5 reposts · 2 replies
Open on X →
Got all that? Now let's look at some real data from biobanks. Again we see that race is a very poor model of human populations, which are continuous mixtures of multiple data sources with no clear boundaries or mapping to any folk racial constructs. https://t.co/fZhkjAcHe1
28.3K views · 118 likes
· 3 reposts · 13 replies
Open on X →
[Lawson 2018] provide many examples where STRUCTURE misses known, uh, structure; merges divergent populations together; finds false admixture; or models ancient genomes as mixtures of modern ones. https://t.co/Pt79KT01v5
15.8K views · 79 likes
· 2 reposts · 2 replies
Open on X →
3: Admixture Graphs use allele-sharing statistics to fit population drift, splits, and mixtures. A very powerful approach but identifiability is a challenge: is the graph you found is significantly better than all other possible graphs? [Maier 2023] show it often is not. https://
15.1K views · 70 likes
· 1 reposts · 2 replies
Open on X →
2: STRUCTURE clusters individuals as mixtures of a fixed set of (k) populations. It shares strengths with PCA (sensitive, interpretable) but also limitations (distorted by sampling and parameter choice). https://t.co/owxvV3nrKU
15.5K views · 66 likes
· 0 reposts · 1 replies
Open on X →
But PCA is easily distorted by the sampling process: bigger populations will warp the PCA locations, even individuals within a single family can look like different populations. PCA also produces unusual artifacts when there is simple spatially locality in the data. https://t.co/
16.9K views · 88 likes
· 1 reposts · 2 replies
Open on X →
1: PCA estimates eigenvectors of the sample relatedness matrix which, under simple ancestry models, are expected to recover population labels. Matrix theory shows that PCA is extremely sensitive with enough data, able to detect relationships down to a handful of generations. http
18.1K views · 95 likes
· 0 reposts · 3 replies
Open on X →
But what about ancestry? Let's go through three core methods for analyzing ancestry in genetic data: dimensionality reduction (PCA), model-based clustering (STRUCTURE), and parametric model (admixture graphs). Because ancestry is relative, each approach has limitations.
18.7K views · 102 likes
· 0 reposts · 1 replies
Open on X →
เทียบกับบัญชีขนาดเดียวกัน
15 โพสต์จาก 90 วันที่ผ่านมา เทียบกับช่วง 10K–100K ผู้ติดตาม แสดงในวงกว้าง แต่มีผู้ชมตอบสนองน้อย.
ยอดดูมัธยฐาน2 964บัญชีนี้924ค่ามัธยฐานของ 10K–100K
การเข้าถึง, %11.81%บัญชีนี้3.62%ค่ามัธยฐานของ 10K–100K
การมีส่วนร่วม, %1.03%บัญชีนี้1.52%ค่ามัธยฐานของ 10K–100K
| ตัวชี้วัด | บัญชีนี้ | ค่ามัธยฐานของ 10K–100K | อัตราส่วน |
|---|---|---|---|
| ยอดดูมัธยฐานต่อโพสต์ | 2 964 | 924 | 3.21× |
| การเข้าถึง (ยอดดู ÷ ผู้ติดตาม) | 11.81% | 3.62% | 3.26× |
| อัตราการมีส่วนร่วม | 1.03% | 1.52% | 0.68× |
บัญชีอื่นในช่วงนี้ → เปรียบเทียบกับบัญชีอื่น → ค่าอ้างอิงเหล่านี้คำนวณอย่างไร →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
26.3K30 Aug
4.3K
25.4K2 Sep
57.6K3 Sep
1K
574
482
7.2K
3K
2.4K
21.6K
1.4K
614
209
Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.83%30 Aug
0.56%
0.62%2 Sep
0.66%3 Sep
1.26%
4.70%
3.94%
0.38%
0.37%
1.26%
0.20%
3.90%
4.23%
4.78%
Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.
What the audience does
Likes81.6%2 628 in total
Reposts3.0%96 in total
Replies5.9%191 in total
Quotes1.4%45 in total
Bookmarks8.0%259 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.
บัญชีที่คล้ายกัน
C.A. ✓
@carimeli
50.2K followers
· 221.5K posts
Dr. Shriram Nene ✓
@DoctorNene
50.2K followers
· 16.1K posts
Senator Leo Housakos ✓
@SenatorHousakos
50.2K followers
· 18K posts
fxhash
@fx_hash_
50.2K followers
· 9.7K posts
Silo Labs | V3 Live ✓
@SiloFinance
50.2K followers
· 6.3K posts
94WIP Afternoons
@AfternoonsWIP
50.2K followers
· 19.2K posts
General Carsten Breuer
@BundeswehrGI
50.2K followers
· 1.5K posts
عبدالله بن ايمن القاضي ✓
@aaalkadi
50.2K followers
· 371 posts
Frequently asked questions about @SashaGusevPosts
How many followers does @SashaGusevPosts have?
25 095 followers.
How many posts has @SashaGusevPosts published?
23 220 posts.
How do posts by @SashaGusevPosts perform?
Collected posts: 36. Views: 1 336 304. Likes: 5 300.