Alex Tong @AlexanderTong7 Twitter profile

Last Seen Profiles

@ofdarksister

@juliemalaure

@seraphemcamille

@Donnell134

@SparkyTheCat_

@aoi_aochan03

@ctk9_m12

@OlmstedSheriff

@_johnnymaga

@pawel_developer

@RyanStevenF2

@wickedalertpod

@loserwbly

@TeuzRoza

@en_tOrihiki_

@pukkapiknewver

@lalilim1

@Roxanne1288895

@lirara_111

@SaudiDesiGirl

@B2B2_B2

@aiAIai_006

@TetraQammon

@Yucha3813

@SM43413

@eeeeaaaaoooowww

@hwuthikwnx71540

@k0koro_091

@GamesLapin

@taniko2505

@SRc94kwb

@harumage_DQX

@SelmaKaraduman

@inazo_game

@HeIsTheCuz

@JoseLGarzon86

Alex Tong

@AlexanderTong7

1 month

Excited to share our new work on Meta Flow Matching (MFM)! 🎉 MFM helps improve the generalization of personalized treatment response predictions by conditioning on learned features of the initial distribution. Work led by @lazar_atan and @NZhang211 .

Lazar Atanackovic

@lazar_atan

1 month

🚀Introducing — Meta Flow Matching (MFM) 🚀 Imagine predicting patient-specific treatment responses for unseen cases or building generative models that adapt across different measures. MFM makes this a reality. 📰Paper: 💻Code:

2

48

267

0

7

79

Alex Tong

@AlexanderTong7

4 months

Excited to share our latest work on Metric Flow Matching (MFM)! Loved working with this team. Amazing to see how far we've come computationally since I started working in this field.

Kacper Kapuśniak

@KKapusniak1

4 months

If data lives on a manifold, how do we design meaningful interpolations between marginals? We present Metric Flow Matching (MFM)… @PPotaptchik @TeoReu @leoeleoleo1 @AlexanderTong7 @mmbronstein @bose_joey @Francesco_dgv 🔗Dive in here: 🧵 (1/12)

9

70

370

0

11

73

Alex Tong

@AlexanderTong7

8 months

Happy to introduce our work on DEM which builds on incredible work on FAB by @SilkyDogfish , Vincent Stamper, @GregorSimms @bschoelkopf and @jmhernandez233 and Equivariant FM by @leonklein26 Andreas Krämer and @FrankNoeBerlin . .

Equivariant flow matching

Normalizing flows are a class of deep generative models that are especially interesting for modeling probability distributions in physics, where the exact likelihood of flows allows reweighting to...

arxiv.org

Joey Bose

@bose_joey

8 months

No data? No problem🚀! Introducing iDEM a new method to sample from Boltzmann distributions. arXiv: co-led with @tara_aksa , @jarridrb and amazing collaborators: @sarthmit @PabloLemosP @ChengHaoLiu1 @MarcinSendera Siamak Ravanbaksh Yoshua Bengio

5

54

242

3

10

70

Alex Tong

@AlexanderTong7

1 year

Check out this extension of DynGFN ( with @lazar_atan and @jasonhartford ) which improves scalability with a more efficient parameterization. Reducing trajectory lengths from n^2 --> n.

DynGFN: Towards Bayesian Inference of Gene Regulatory Networks...

One of the grand challenges of cell biology is inferring the gene regulatory network (GRN) which describes interactions between genes and their products that control gene expression and cellular...

arxiv.org

Dianbo Liu

@DianboLiu

1 year

Curious about how to discover causal interactions among many genes in a cell with uncertainty estimation? @TrangNguyen2399 @AlexanderTong7 Check our new work :

1

11

57

0

5

48

Alex Tong

@AlexanderTong7

1 year

Happening in ~2 hours! Code available here:

GitHub - atong01/conditional-flow-matching: TorchCFM: a Conditional Flow Matching library

TorchCFM: a Conditional Flow Matching library. Contribute to atong01/conditional-flow-matching development by creating an account on GitHub.

github.com

Hannes Stärk

@HannesStaerk

1 year

Diffusion models are dead - long live joint conditional flow matching! 🙃 Tomorrow @AlexanderTong7 presents his "Improving and generalizing flow-based generative models with minibatch optimal transport" On Zoom 11am EDT / 3pm UTC:

4

56

336

0

8

48

Alex Tong

@AlexanderTong7

1 year

Had a great time working with this team. See our first work on flow matching for protein generation. More to come!

Joey Bose

@bose_joey

1 year

What's better than Diffusion for protein backbone generation? (Stochastic) OT Flow Matching over SE(3)! link: New paper w/ @tara_aksa @FatrasKilian @Guillaume_hu @jarridrb @ChengHaoLiu1 , @andreic_nica @MaksymKorablyov @mmbronstein @AlexanderTong7 1/5🧵

5

59

335

0

5

39

Alex Tong

@AlexanderTong7

1 year

Very cool application of Flow matching! Multi-Ligand binder design by learning flows from a prior with separated ligands and target.

Hannes Stärk

@HannesStaerk

1 year

New paper!🤗 "Harmonic Self-Conditioned Flow Matching for Multi-Ligand Docking and Binding Site Design" Generating pockets to bind small molecules has applications like designing antidotes that sequester toxins or as a first step toward Enzyme design 1/11

2

64

327

1

4

38

Alex Tong

@AlexanderTong7

2 months

Looking forward to presenting this work with @Guillaume_hu tomorrow!

Machine learning for protein engineering seminar

@ml4proteins

2 months

This week, 8/06 @ 4 pm ET, we'll have @AlexanderTong7 present Sequence-Augmented SE (3)-Flow Matching For Conditional Protein Backbone Generation Read about FoldFlow FF2: FF1: Sign up to receive the Zoom link via email!

1

14

63

0

6

36

Alex Tong

@AlexanderTong7

4 months

Checkout our new protein model @DreamFoldAI thread by first author @Guillaume_hu . Great work by the team and proud of our results!

Guillaume Huguet

@Guillaume_hu

4 months

Introducing FoldFlow-2, a new SOTA sequence-conditioned protein generative model! work w/ @DreamFoldAI James V @FatrasKilian Eric TL @PabloLemosP @riashatislam @ChengHaoLiu1 @jarridrb @tara_aksa @mmbronstein @AlexanderTong7 @bose_joey Arxiv: 1/8 🧵

4

53

216

0

3

33

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z Optimal transport is incredibly useful in single-cell analysis, but how do you handle controls? @christophertape already has a great thread on the biological impact, so in this thread I’ll focus on the ML and computational aspects. (1/n)

1

6

29

Alex Tong

@AlexanderTong7

1 year

Thanks for having me! Had a great time at @HelmholtzMunich and many inspiring discussions.

CausalCellDynamics_International Lab

@causal_cell

1 year

Thank you @AlexanderTong7 from @Mila_Quebec for having stopped over at @HelmholtzMunich and sharing your current work perspectives; Looking forward to your come-back, to enjoy more discussions on cell causality (and the famous Munich beach) (+thanks @THJimmyLee for the photo)

0

12

1

0

16

Alex Tong

@AlexanderTong7

10 months

I wholeheartedly agree😀! Great working with @christophertape and @Maria_Ramos_Z on this!

Chris Tape

@christophertape

10 months

It has been such an enormous privilege to work with @KrishnaswamyLab and @AlexanderTong7 on this project. Alex is currently on the faculty job market and anyone who has any sense should hire him immediately.

1

3

36

0

15

Alex Tong

@AlexanderTong7

2 years

Great tutorial on Trellis created by @Maria_Ramos_Z which shows how to combine hierarchical clustering and optimal transport for compare a large number of single-cell distributions.

Maria Ramos

@Maria_Ramos_Z

2 years

Here it is! We have now created a how-to tutorial to understand and use Trellis, a tool for hierarchical clustering that enables the fast calculation of distances across thousands of distributions containing single-cell information:

1

5

26

0

1

14

Alex Tong

@AlexanderTong7

3 years

Happy to be defending my thesis in a few hours! If you want to hear about our work in graphs, optimal transport, and deep learning, hope to see you there!

0

11

Alex Tong

@AlexanderTong7

9 months

Awesome blog posts on the state of the field by @michael_galkin and @mmbronstein !

Michael Galkin

@michael_galkin

9 months

📣Two new blog posts - a comprehensive review of Graph and Geometric ML in 2023 with predictions for 2024. Together with @mmbronstein , we asked 30 academic and industrial experts about the most important things happened in their areas and open challenges to be solved. 🧵 1/n

3

129

457

0

1

12

Alex Tong

@AlexanderTong7

9 months

Great working with this incredible team! The most recent version (with updated results) is available on the open review site here:

SE(3)-Stochastic Flow Matching for Protein Backbone Generation

The computational design of novel protein structures has the potential to impact numerous scientific disciplines greatly. Toward this goal, we introduce \foldflow, a series of novel generative...

openreview.net

Joey Bose

@bose_joey

9 months

Happy to share that FoldFlow, the first Flow Matching generative model for Protein backbone generation got accepted to #ICLR2024 as a Spotlight with scores 8/8/8/8! Thanks to the reviewers for the feedback and ofc all my incredible co-authors for pulling this fun paper off🚀!

4

16

130

0

11

Alex Tong

@AlexanderTong7

10 months

Hope to see you there! TL;DR GflowNets for Bayesian (Dynamic) Gene Regulatory Network discovery.

Lazar Atanackovic

@lazar_atan

10 months

I will be at #NeurIPS2023 this week! I am excited to be presenting DynGFN during the main conference. Looking forward to seeing everyone there! Poster session 4, Wed, Dec 13th 5 - 7 pm CST @VectorInst @Mila_Quebec @valence_ai

0

6

23

0

1

10

Alex Tong

@AlexanderTong7

9 months

Can't recommend working with @christophertape and lab enough! An awesome opportunity.

Chris Tape

@christophertape

9 months

1 week left to apply for our computational PhD project 'Single-cell Signalling Analysis of Stem Cell Polarisation in Cancer'. 4-years, fully funded, co-supervised across @uclcancer @crick with a placement at @Yale Pls RT!

0

24

22

0

1

8

Alex Tong

@AlexanderTong7

8 months

@jonkhler @SilkyDogfish @GregorSimms @bschoelkopf @jmhernandez233 @leonklein26 @FrankNoeBerlin @smnlssn @__kraemer__ Thanks Jonas for your thoughts! I believe your understanding is correct. When we test on the 55 particle Lennard-Jones system we get forces on the order of 10^30. In practice we clip and it seems to work fairly well. Definitely interested in how well it extends to proteins :)

1

0

6

Alex Tong

@AlexanderTong7

1 year

@zdhnarsil

InstaFlow: One Step is Enough for High-Quality Diffusion-Based...

Diffusion models have revolutionized text-to-image generation with its exceptional quality and creativity. However, its multi-step sampling process is known to be slow, often requiring tens of...

arxiv.org

1

0

6

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z Thank you for reading! Code: Tutorial: Raw data: Preprocessed data: (10/n)

GitHub - MariaRamosZ/Trellis_how_to: This is a step by step tutorial on how to implement Trellis in...

This is a step by step tutorial on how to implement Trellis in a training dataset. - MariaRamosZ/Trellis_how_to

github.com

0

1

6

Alex Tong

@AlexanderTong7

1 year

@alexechu_ @bose_joey @json_yim Hi @alexechu_ , I was also curious about this. (1) It turns out that FM and diffusion have slightly different conditional paths as shown in Figure 3 of . (2) the conditionals for the VE ODE are an optimal transport map for a linear noise schedule.

Flow Matching for Generative Modeling

We introduce a new paradigm for generative modeling built on Continuous Normalizing Flows (CNFs), allowing us to train CNFs at unprecedented scale. Specifically, we present the notion of Flow...

arxiv.org

1

0

4

Alex Tong

@AlexanderTong7

8 months

Great feature by @kevanchu and @dow_lab on our work on Trellis studying the Tumor microenvironment in PDOs.

Krishnaswamy Lab

@KrishnaswamyLab

8 months

Check out this Cancer Research feature on our recent work on PDOs and Trellis with @christophertape @AlexanderTong7 @Maria_Ramos_Z !

0

3

11

0

4

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z I’m excited to see what types of methods can be developed using this data. High-level summary, 25m cells, 44 features in a tensor of 10 patients, 11 drug combinations, 3 co-cultures states, with 3 replicates, which presents many opportunities for splitting and validation. (9/n)

1

0

3

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z Single cell datasets are getting larger and larger. Heren @maria_ramos_z profiled >25 million cells in 3,360 separate samples spanning patient, drug, and microenvironment specific effects. At this scale it’s difficult to understand the data let alone draw insights. (2/n)

1

0

2

Alex Tong

@AlexanderTong7

2 months

@Pseudomanifold @unifr Congrats Bastian

1

0

2

Alex Tong

@AlexanderTong7

4 years

Grace choreographs a Tod's Shoes commercial via @richtong

OMG Heart swells with pride, Grace choreographs a Tod’s Shoes commercial

Well, what a cool thing, even in pandemic/election times, Grace choreographs and performs in a commercial that is tribute to New York City for Tod's Shoes. Got

tongfamily.com

1

0

2

Alex Tong

@AlexanderTong7

6 months

@k_neklyudov @UMontreal @Mila_Quebec Congrats Kirill!

0

2

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z We developed Trellis as an extremely fast and interpretable method to understand how a treatment changes the observed cell states. Trellis is able to measure the changes in changes of cell state distributions by how much work it takes to transport over a cell state tree. (6/n)

1

0

2

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z The L^1 distance between these vectors is now the Wasserstein distance. This means that when used to create manifold-based embeddings between distributions which only require nearest neighbor distributions, Trellis scales log-linearly in the number of cells and samples. (8/n)

1

0

2

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z Since this dataset is so large, it was collected using many individual experiments. This leads to technical batch effects. What saves us here is that each experiment has its own controls. (3/n)

1

0

2

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z Trellis embeds distributions to vectors where the Wasserstein distance between distributions is equivalent to the L^1 distance between vectors. On a tree with hierarchical clustering, we compute the cell abundances in each node of the tree weighted by the edge weight. (7/n)

1

0

2

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z This means that we can greatly reduce technical variation by comparing how treatments affect the cell distribution within the same plate. (5/n)

1

0

2

Alex Tong

@AlexanderTong7

10 months

@KrishnaswamyLab @CellCellPress @christophertape @Maria_Ramos_Z Since this dataset is so large, it was collected using many individual experiments. This leads to technical batch effects which can drown out subtle biological signals. What saves us here is that each individual experiment has its own controls. (4/n)

1

0

2

Alex Tong

@AlexanderTong7

3 months

@yuanzhi_zhu @iScienceLuvr Yes! Cool to see what we call the minibatch-OT trick also works in these settings! Some works that may interest the authors ( @Yiheng_Li_Cal @Chenfeng_X ) that also use this trick: OT-CFM Multisample-FM A-OT

1

0

2

Alex Tong

@AlexanderTong7

2 years

@KrishnaswamyLab @YaleMed @Yale @YaleCompsci @YaleGenetics Congratulations!

0

1

Alex Tong

@AlexanderTong7

2 months

@Pseudomanifold @unifr Would love to :) Hopefully I'll make it at some point.

0

1