gokul gowri

learning about living things @ mit eecs + whitehead institute

[gokul gowri @gokl_g]

@arshia_hemmat @erturklab @njwfish @WilliamNColgan fusion events are defintely beyond the scope of the model, but the asymmetric splits were something we thought about a lot. we generate layers of the tree simultaneously w a transformer so we can capture sister-dependence :) we show this in the synthetic data exps in fig 2

[gokul gowri @gokl_g]

@kdpisda @JswLab @njwfish @WilliamNColgan @stats_stephen @marinkazitnik @LukeKoblan it’s a good question :) we didn’t explicitly test this, but fig 4e of the preprint strongly suggests the memory in neurons is stable.

[gokul gowri @gokl_g]

the transcriptional state of a cell is not generally sufficient to predict its past or future. in our new preprint we show that conditioning on lineage context (sisters, cousins, etc) can help a lot.

this one was v fun :)

https://t.co/nmlsV3o1fP

[Jonathan Weissman's Lab @JswLab]

New preprint on generative modeling for mouse development! Rather than single cell trajectories, we set out to model embryos as trees.

Led by @gokl_g, @njwfish, @WilliamNColgan, w/ Philippe Rigollet, @stats_stephen, Zack Smith, @marinkazitnik, @lukekoblan and a wonderful team

[Nic Fishman @njwfish]

Finally out! We develop an approach to learn dynamics over lineage trees (w/o OT assumptions) and built a diffusion model to sample those dynamics.

We show the model can identify how initially different cells converge to the same fate and can even predict perturbation effects.

[Jonathan Weissman's Lab @JswLab]

New preprint on generative modeling for mouse development! Rather than single cell trajectories, we set out to model embryos as trees.

Led by @gokl_g, @njwfish, @WilliamNColgan, w/ Philippe Rigollet, @stats_stephen, Zack Smith, @marinkazitnik, @lukekoblan and a wonderful team