/Tech6h ago

Embeddings Reveal Correlations But Do Not Guarantee Causal Biology

21801720
Anshul Kundaje#1779
Original post
Anshul Kundaje@anshulkundaje#1779inTech

You can always pull out anecdotes from your models representation where some correlation structure reveals causal biology. Eg. Co-expression networks certainly carry nuggets of causal relationships but there are no guarantees. This doesn't make the model causal. 4/

Anshul Kundaje@anshulkundaje

It's worth noting that predictive representations (including embeddings from scFMs) can be useful to learn causal models. But without the right data inputs & expt design, it's literally impossible to magically learn biologically & statistically causal models. 3/

3:23 PM · Jun 10, 2026 · 369 Views
Sentiment

Positive users value large models that reliably deliver promised results, while negative users insist such models must clearly outperform simpler approaches on meaningful outcomes.

Pos
100.0%
Neg
0.0%
2 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS351BOOKMARKS1LIKES9REPLIES1
Anshul Kundaje@anshulkundaje

However, one can solve many problems eg cell type annotation, batch correction, perturbation prediction (within specific limits around the training/fine tuning data) without ever learning a causal model. Simple approaches can do extremely well when provided the right data. 5/

Anshul Kundaje@anshulkundaje

You can always pull out anecdotes from your models representation where some correlation structure reveals causal biology. Eg. Co-expression networks certainly carry nuggets of causal relationships but there are no guarantees. This doesn't make the model causal. 4/

6hViews 351Likes 9Bookmarks 1
Anshul Kundaje@anshulkundaje

If u are building a large model, better make sure it actually reliably trashes the simpler approaches on outcomes that matter or offers unique insights & make sure it's really worth the bang for the buck. Nobody is going to use huge models for marginal or no gains. 6/

6hViews 66Likes 2
Anshul Kundaje@anshulkundaje

On the other hand, if u provide a model that consistently delivers what it promises, everyone will keep their mouths shut & happily use it. Quite easy to make the case. The model should be able to speak for itself. 7/7

6hViews 63Likes 3