Aditi Qamra

Computational biologist | Cancer researcher. Ph.D. @astar_gis Postdoc @UHN. Early drug development and biomarker discovery @RocheCanada. Now @arteraAI

[Aditi Qamra @Itti_Q]

I want to see models built on mechanistic biological data. There is abundant experimental data and literature which has not been consolidated to build a well informed updated picture of the human cell 1/

[Marios Georgakis @MariosGeorgakis]

It indeed feels like the entire "AI×bio" industry is under a massive illusion about the hurdles of drug development.

I wonder how long people will continue raising billions on the basis of "we'll run AI on all scientific literature and multi-omics data, and we'll cure all disease in 10 years"

This is really a great piece.

[Aditi Qamra @Itti_Q]

@david_van_dijk Appreciate the intent of getting high quality data and training across multiple assays. Why a LLM like model though? Thanks!

[Aditi Qamra @Itti_Q]

My mum, maternal aunts and grandma all have AMD. High prob I will too. a-VEGF injections did not prevent their vision loss. This gives me a lot of hope

[Max Hodak @maxhodak_]

Big news today. Science's PRIMA visual prosthesis, our retinal implant for macular degeneration, has received its CE mark, which means marketing approval in Europe. This is a huge milestone to begin to have a device like this commercially available to regular patients. 1/

[Aditi Qamra @Itti_Q]

@pdhsu Such as ?

[Jake Wintermute 🧬/acc @SynBio1]

Everyone wants to cure all diseases but nobody wants to do no hard ass biology experiments

[Shiv Grewal @grewalsh]

Introns have a hidden regulatory role! 🧬🎉
Delighted to share our latest paper showing that inefficiently spliced introns and spliceosomal proteins direct RNA methylation, engaging RNAi to silence retrotransposons and regulate gene expression
https://t.co/tVTWTewauk

[Martin Borch Jensen @MartinBJensen]

To be clear, not saying sc is a bad tool.

But 'predicting novel targets' means engaging with pathophysiology you're trying to reverse. Cell culture doesn't have pathophysiology.

The models aren't built for novel targets in complex disease. Good work, for a different question.

[Aditi Qamra @Itti_Q]

This this this

[Yusuf Roohani @yusufroohani]

How do we make experimental velocity the metric that determines which AI-for-biology models have the greatest impact?

scaling data < scaling AI-guided experiments

I'm at ICML 🇰🇷, let's chat!

[Aditi Qamra @Itti_Q]

Its the first im seeing of experimental validation being treated at equal pedestal as drug design. Huge kudos to the team for baking it in their pipeline and then optimizing for efficiency rather than as an afterthought

[Isaak @isaakfreeman]

How did we achieve so fast end-to-end testing? Our proprietary Capable Core: We verticalized key steps of preclinical development. Our platform consists of 6 semi-autonomous modules, most of which we re-engineered to run 10–20x faster.

Drug design, ML infra, rapid synthesis, quality control, cell testing, and then mouse testing.

Our first modality at the 24h pace is binders & peptides--including complex modifications, cyclization, and noncanonical amino acids.

Mouse studies: For many of the biologically fast readouts, such as our short sleeper programs, we are able dose new drugs in placebo-controlled in vivo studies within that 24h period too and can read out early significant differences between drug candidates rapidly.

We achieved this through an in vivo facility in the lab, pre-approved protocols, and veterinarians & in vivo researchers on standby. This doesn't work for all indications: For Alzheimer's models the time-to-first-signal in vivo read-out is longer. For all drug candidates, before clinical testing we do standard chronic studies.

[euan ashley @euanashley]

Our recommendations include:

→ Transition to genome benchmarking

It is time to move beyond isolated small-variant metrics towards evaluating entire genomes against complete, diploid reference sequences to accurately evaluate and benchmark complex genetic variation.

[Jason Locasale @LocasaleLab]

As an AI optimist, I do think AI will fundamentally transform biology and medicine. But after 25 years in academia and biotech and doing biology and quantitative biology at the highest levels, I think many AI companies encounter two very hard truths as tech companies have in the past.

First, drug development is constrained by regulatory requirements, clinical trial frameworks, and their incentive structures that determine what can be discovered, tested, and ultimately brought to patients. Better models alone do not solve those bottlenecks.

Second, AI is only as powerful as the questions it is asked. Biology is an extraordinarily difficult field because we still lack the right conceptual frameworks for understanding how living systems work. AI will accelerate discovery, but only if someone knows which problems matter, what data are informative, and how to distinguish mechanism from correlation.

In my experience, people who can do that are incredibly rare. Many areas of biology have only a handful of scientists with both deep biological intuition and the mindset needed to fully exploit AI. They know which data should exist but don’t, which experiments will change our understanding, and which questions are worth asking in the first place.

AI companies recruit elite computer scientists with extraordinary urgency. If they want to transform biology, they will need to recruit elite biologists with the same urgency. Otherwise, they’ll be limited by the quality of the biological questions those models are asked to solve.

[Ran Blekhman @blekhman]

Claude Science is incredible. I gave it some sequencing data, and in 8 hours it did a full analysis, generated figures, wrote a paper, submitted it for publication, got rejected, revised and resubmitted, got rejected again, it is now applying for positions in industry

[Aditi Qamra @Itti_Q]

@tangming2005 What sources are you using to learn? Im trying to as well so im curious to share notes

[Anshul Kundaje @anshulkundaje]

Hope "AI Scientists" don't go the same trajectory as many bioinformatics tools. 10000 flavors. Impossible to know who is better for what purposes. Everyone benchmark maxxing. Users bewildered & confused. Hopefully they will all be equally useful rather than equally useless.

[Aditi Qamra @Itti_Q]

So friggin cool. I love the fact that this was achieved non-traditionally. Started out of a lab set up in an airbnb, scientists with lateral expertise came together, upskilled and achieved this vision - huge kudos 👏

[Aleph @alephneuro]

We recently obtained the highest-resolution 3D images of the human brain ever taken from outside the skull. This is the first look.

Introducing Aleph, a research lab building brain interfaces for the telepathic future. (1/n)

[Aditi Qamra @Itti_Q]

@jkobject That would be great! I should say while gene regulation is my area of expertise, building self learning models is not. Having said that, imo the 1st step is identifying high-quality literature/experimental datasets. I can start to draft some prelim thoughts

[Aditi Qamra @Itti_Q]

If we are talking about pipe dreams re imaging then here is mine https://t.co/jmAnT0Qe3k

[Martin Pacesa @MartinPacesa]

Shows very nicely why the current “virtual cell” efforts are just a waste of compute. People throwing data at models hoping that they learn principles without the proper context, rather than putting effort in to actually capture the underlying biochemistry/biology.

[Aditi Qamra @Itti_Q]

@jkobject Oh wow! What a lovely find and coincidence - I was mulling over consolidating all known mechanistic data into one map just a while ago (https://t.co/dPnHk7VuTD) Looking fwd to digging into resource further!

[Aditi Qamra @Itti_Q]

I want to see models built on mechanistic biological data. There is abundant experimental data and literature which has not been consolidated to build a well informed updated picture of the human cell 1/

[Aditi Qamra @Itti_Q]

It would enable better design of small molecules. Pairing with spatial and/or histopathology data for cellular microenvironment will help answer the why behind response heterogeneity

[Aditi Qamra @Itti_Q]

I want to see models built on mechanistic biological data. There is abundant experimental data and literature which has not been consolidated to build a well informed updated picture of the human cell 1/

[Aditi Qamra @Itti_Q]

Chasing phenotypic changes for clinical impact is a low hanging fruit but I believe we need a systemic upto date map of the machinery re transcription, signaling pathways and molecular switches in different cell types, disease states, and developmental stages 2/