Kexin Huang
@KexinHuang5
Followers
1,823
Following
562
Media
50
Statuses
367
PhD Student @Stanford CS with @jure ; Machine Learning + Biomedicine
Joined August 2018
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
オーロラ
• 492250 Tweets
Joost
• 334472 Tweets
#الهلال_الحزم
• 85607 Tweets
Fulham
• 81859 Tweets
#SixTONESANN
• 78237 Tweets
Man City
• 62845 Tweets
自分これ
• 61859 Tweets
Gvardiol
• 54232 Tweets
Ali Koç
• 47760 Tweets
Dremo
• 41197 Tweets
太陽フレアのせい
• 39467 Tweets
Burnley
• 33746 Tweets
SEE YOU NEXT MONTH JIN
• 31685 Tweets
دوري روشن
• 30597 Tweets
الدوري الاقوي
• 28769 Tweets
Granada
• 25210 Tweets
روشن السعودي
• 24591 Tweets
Luton
• 22315 Tweets
Haley
• 22196 Tweets
やまとなでしこ
• 14763 Tweets
الدوري التاريخي
• 14541 Tweets
#الهلال_بطل_الدوري_التاريخي
• 11927 Tweets
1/🧵Introducing Perturb-seq-in-the-loop: a sequential experimental design strategy for perturbation screens guided by multimodal priors, with 3X speedup over state-of-the-art active learning methods!
With amazing
@_romain_lopez_
@jchuetter
Taka Kudo
@antonio_science
Aviv Regev
7
113
473
An update ➡️ I will be attending
@Stanford
CS PhD program in the fall!! Super excited and grateful🙏🙏 Look forward to continuing research on machine learning + biomedicine!!
15
12
304
📣Excited to share our work on conformalized GNN with
@YingJin531
, Emmanuel Candes and
@jure
!
Given an entity in the graph, it produces a prediction set/interval that provably contains the true label with pre-defined coverage probability (e.g. 90%):
3
57
222
Excited to share our work on graph ML to model multiple and never-before-seen genetic perturbations! Tweetorial⬇️
Can cellular response to the perturbation of multiple genes be predicted? What if those genes were never perturbed experimentally?
Yes! We present GEARS, a geometric deep learning model that predicts novel multi-gene perturbation outcomes 1/7🧵
Preprint:
6
80
350
0
20
167
Excited to share a survey on ML for therapeutics tasks centering around genomics data
@Patterns_CP
!
We review 22 tasks across Tx pipelines that study the interplay of DNA seq, omics, compounds, proteins, texts, networks, and spatial data
4
30
151
Excited to share TxGNN! We study in-depth how to make a knowledge graph system for therapeutic use predictions actionable and practical.
🧵1/7
3
28
101
Conformalized GNN to produce reliable graph predictions with statistical guarantees is now accepted by
#NeurIPS2023
as a spotlight!
Code is now available at
📣Excited to share our work on conformalized GNN with
@YingJin531
, Emmanuel Candes and
@jure
!
Given an entity in the graph, it produces a prediction set/interval that provably contains the true label with pre-defined coverage probability (e.g. 90%):
3
57
222
1
11
100
We are hosting a
@LogConference
Bay Area local meetup on Nov 29th! Come to Stanford to learn from and connect with the local graph learning community! Register here:
with
@ShirleyYXWu
@MinkaiX
@jure
0
24
93
Super excited to share our work on predicting unseen & combinatorial perturb-seq outcome using prior knowledge GNN
@NatureBiotech
lots of new results in the updated paper - check them out!
📢 Exciting News! Our latest paper is now out in Nature Biotech 🌱🧬 We developed GEARS---an AI method to predict cellular responses to genetic perturbation. 🧪🔬
🔗 Link to the paper:
🧬 Unraveling genetic interactions in cancer, regenerative medicine,…
3
122
579
4
12
88
Happy to share SumGNN in Bioinformatics!
We inject biomedical knowledge from Hetionet KG and provide a scheme to generate potential mechanism pathways! Great results on DDI!
Paper:
GitHub:
Great work led by
@yueyu30308379
0
12
82
MolDesigner is in
#NeurIPS2020
Demo!
-Interactive molecule design with DL, powered by DeepPurpose and
@GradioML
!
-Predict binding affinity and 17 ADMET properties from 50+ DL models!
-Less than 1 sec latency!
Video:
Paper:
5
19
80
DeepPurpose is in Bioinformatics! ()
A scikit-learn style library for DTI, DDI, PPI, Drug Property/Protein Function Prediction with 15+ DL models
A demo on getting 25 ADME MPNN models with ~10 lines of code, with TDC ⬇️
Github:
2
11
70
Thanks Ahmed for having me and really enjoyed the discussion! Here is also the link for the slide:
In our 3rd Conformal Prediction seminar,
@KexinHuang5
discusses his work w/
@YingJin531
, Emmanuel Candès, and
@jure
, exploring the application of conformal prediction to graph neural networks!
Watch the talk at:
0
9
64
0
5
69
We are hosting a
@LogConference
bay area local meetup! Come to Stanford to learn about exciting new research in LOG and hang out with graph ML folks! Register here:
with
@jure
@MinkaiX
@kaidicao
Lata Nair
0
11
69
I will be at NeurIPS presenting ⬇️ on Thursday poster session and also co-organizing
@AI_for_Science
workshop on Saturday! PM me if you would like to chat about anything related to AI + biological discovery!
📣Excited to share our work on conformalized GNN with
@YingJin531
, Emmanuel Candes and
@jure
!
Given an entity in the graph, it produces a prediction set/interval that provably contains the true label with pre-defined coverage probability (e.g. 90%):
3
57
222
1
2
59
Very happy to share our perspective on the exciting intersection of graph ML + biomed & healthcare
@natBME
!
🙌We are beyond excited to share our Perspective on graph representation learning for biomedicine and healthcare!🥳
@natBME
@KexinHuang5
@marinkazitnik
@HarvardDBMI
(1/4)
6
37
214
0
7
51
How to do node/link prediction when only a handful of labels are available? SOTA GNNs would fail!
Come to poster B1-C1 on Thursday 9pm PT
#NeurIPS2020
! ()
Project website: with
@marinkazitnik
!
0
15
46
Very excited to share Therapeutics Data Commons, an ML data hub for therapeutics!
3 lines of code (!) to access
- 62 ML-ready datasets from 22 Tx tasks
- 10 Biologics Datasets
- 24 ADMET Datasets
- 20 Mol Oracles
Talks tmr 1:45pm EST at
#futuretx20
!
1
14
42
14/ There are still lots of open questions such as how to extend to combinatorial perturbations, how to simulate batch effects better, etc. We are excited for this direction!
Preprint:
Code:
Talk at MLCB:
2
6
45
Excited to share our perspective on AI + Science! We review how AI accelerates the scientific inquiry loop (hypothesis -> experiments -> observations)
@AI_for_Science
How can
#AI
transform science?
Let us count the ways
A brilliant review
@Nature
@marinkazitnik
@TianfanFu
@YuanqiD
and colleagues
@AI_for_Science
#ScienceTwitter
12
104
330
0
5
42
Many international students don’t have access to resources on how to do proper research.
To fill the gap, with
@jimeng
and
@caoxiao_danica
, we initiate “Sunstella Mentorship Fellow” for PhD/PostDoc/… who are excited to help international junior students over summer 2022⬇️
2
8
39
Awesome blog about exciting graph ML works in the past year! Also checkout my reviews on the graph learning for system biology and robustness sections!
📣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
449
0
4
32
Excited to share our annual review on 🔥 research in AI for science from 2023!
🚀 Exciting News! Our blog “AI for Science in 2023: A Community Primer” is now live! In this blog, we delve into how AI intersects with various scientific fields - from Chemistry, Biology, Physics, Computer/Math. Science, Neuroscience to Earth Science.
2
33
109
1
1
32
Super excited to share TDC paper
@nchembio
- see tweetorial ⬇️
Excited to share our new paper in Nature Chemical Biology
@nchembio
AI is poised to transform
#therapeutic
#science
The Commons is an initiative to access and evaluate
#AI
capability across therapeutic modalities and stages of discovery 1/4
1
19
72
1
5
31
Happy to share the first leaderboard of TDC: 22 new & important ADMET property prediction datasets for molecular ML model! Submit your model to TDC!
More info:
GitHub:
@marinkazitnik
@TianfanFu
@WenhaoGao1
@yzhao062
@yusufroohani
0
9
28
Checkout MolTrans, a new deep learning model for Drug-Target Interaction Prediction! Now published in Bioinformatics!
Code:
Paper:
Check out our paper of a new deep learning method for drug target interaction (DTI) prediction
@AiLucasg
@caoxiao_danica
@KexinHuang5
1
3
17
3
4
24
Excited to co-organize this workshop⬇️
Introducing AI for Science — a
@NeurIPSConf
2021 Workshop! Our workshop focuses on bridging the gaps between machine learning and science. We have a stellar lineup of speakers! Submit your work and sign up for mentorship program now:
2
52
189
0
1
24
Just create a web UI for Drug-Target Interaction Prediction with less than 10 lines using DeepPurpose and
@GradioML
👇 Very impressed with Gradio's simplicity!
2
10
21
Excited to share G-Meta with
@marinkazitnik
:
- Local subgraph enables knowledge transfers in various graph meta-learning problems, with theoretical motivation!
- Promising result on 7 datasets, where two are new & large (1,840 graphs)!
- G-Meta scales!
-
0
6
21
Our new preprint on a comprehensive survey for graph learning + biomedicine! Lots of exciting opportunities: Link ⬇️
I am excited to finally announce our survey of representation learning for networks in bio. and med. w/
@KexinHuang5
&
@marinkazitnik
! We synthesize a spectrum of rep. learning approaches & share 4 unique prospective studies to demonstrate their potential!
1
8
38
0
1
21
How to extract a clinical variable out of the entire patient’s history notes of > 200K words, using an expensive BERT-style model?
Happy to share SnipBERT done
@flatironhealth
in NeurIPS ML4H!
- up to >20% prediction gain
- clues for model interpretation
Details ⬇️
2
4
19
Excited to co-organize AI for Science workshop again at NeurIPS 2023!
(1/3) I am thrilled to announce that
@AI_for_Science
workshop is back with
#NeurIPS2023
! This year we put together several new programs with a new theme "from theory to practice", including a panel discussion to align the expectation between academia and funding agencies.
3
36
144
0
0
18
New TDC Leaderboard on docking score molecule generation⬇️
We restrict the Oracle calls to simulate realistic wet-lab constraint; realistic eval metrics such as %pass mol filters and
@MoleculeOne
for synthesizablity! => Most SOTA fail!
Happy to announce our latest
#benchmark
and
#leaderboard
on molecular
#docking
to evaluate approaches for structure-based
#drugdiscovery
and
#molecule
generation
0
9
28
0
1
14
TDC preprint is alive in arXiv! ⬇️
Excited to share preprint on Therapeutics Data Commons!
Paper:
Website:
TDC is a unifying framework across the entire range of
#therapeutics
#ML
. Ecosystem of tools, leaderboards & community resr
66 ML-ready datasets
22 ML tasks
1
62
220
0
2
14
Really excited for my first NeurIPS paper! We introduce a general and effective framework for graph meta learning, check it out: . Thanks Marinka for the great guidance!
Thrilled that our lab has 4/4 papers accepted at
#NeurIPS2020
! Not bad for a lab just 5 months old at submission deadline. Congrats to fantastic students and collaborators
@_michellemli
@xiangzhang1015
@KexinHuang5
@IAmSamFin
@Emily_Alsentzer
@Harvard
@HarvardDBMI
@harvard_data
11
7
182
2
1
14
2/ 🧬Perturbation screens can answer central questions in biomedicine but usually require large-scale profiling of perturbations across various cellular contexts, surpassing the capacity of the largest facilities.
1
0
14
A tutorial of DeepPurpose for data scientists👇
@TianfanFu
@AiLucasg
@marinkazitnik
@caoxiao_danica
@jimeng
@IQVIA_global
@harvard_data
Drug Discovery with Deep Learning Under 10 Lines of Codes by Kexin Huang
0
6
26
0
2
12
LoG is happening tomorrow! Join now:
LoG is happening tomorrow!
Highlights of the program:
🎤Exciting keynotes from
@jure
,
@andreasloudaros
, Stefanie Jegelka,
@KyleCranmer
,
@ktschuett
🌟 12 orals
💻 Tutorials on scalability & recommendation
🤗 poster sessions & networking
Join now via
1
32
104
0
0
11
6/ AL is typically used for large samples with many rounds, but our economic analysis shows that each round of perturb screens takes ~1 month and each perturb costs ~$30. So we can only do very few rounds and also a few perturbations. We call this “active learning on a budget”.
2
0
10
3/ ML models have been proposed to predict unseen perturbations. However, it is hard to generalize to the entire perturb space because current models are trained on perturbations designed by biologists to answer specific questions, but not to explore the whole perturbation space.
1
0
10
10/ Our key method is to map every modality into the same kernel space and guide the update of the model kernel - leading to a more accurate kernel that captures perturbation relations, and in turn, a better selection strategy.
1
0
9
11/ Our method works well compared to uniform sampling, with 5X speedup but also has competitive (3X!) speedup over SOTA AL methods. Notable gain occurs in the first two rounds!
1
0
9
4/ What if we can ask the model to select the most needed perturbations? This way we can do as little experiments as possible. We call this iterative perturb-seq where we let the ML model select the next batch of perturbations that are used to sequentially improve the model!
1
0
10
Very happy to work with and learnt a ton from
@YingJin531
Emmanuel and
@jure
on this project.
Guaranteed uncertainty has extensive applications in graph ML, biomedical ML, drug discovery ML, that’s why I am excited about it! Stay tuned for our future work in this direction!
1
1
8
We just released four new realistic oracles on docking scores and synthetic pathway analysis from ASKCOS,
@MoleculeOne
, and
@ForRxn
, all under one-line-of-code!
“The current evaluations for generative models do not reflect the complexity of real discovery problems.”
TDC now includes one-liners for four more realistic oracles: docking scores, synthetic accessibility from ASKCOS,
@MoleculeOne
,
@ForRxn
.
Info:
0
7
23
0
2
8
13/ We also performed many additional experiments such as ablations with individual priors, genome-scale perturb-seq data, etc. (check out the paper!)
1
0
8
12/ As each round consists of a different screen, we identify batch effect as a key real-world challenge. We simulate the batch effect by restricting the model in each round to only access cells in distinct 8 lanes/batches. We still observe consistent improvement.
1
0
8
8/ Our key insight is that there is rich prior information about relations between perturbations including optical pool screens (OPS), PPI, literature, protein emb, other perturb-seq etc. For example, perturbations that elicit similar morphologies in OPS have similar expressions.
1
0
9
5/ This falls into batch-mode active learning (AL) regime, where it selects based on (1) informativeness (2) diversity, and (3) representativeness. We use a recent framework that summarizes all SOTA AL with kernel formulation, where kernel defines relations between perturbations.
1
0
8
Super happy to see this preprint out! Lucky to work with and learned a ton from amazing collaborators
@marinkazitnik
@payal_chandak
@WangQianwenToo
Shreyas, Akhil,
@jure
@girish_nadkarni
@ngehlenborg
7/7
0
0
7
Conformal prediction requires exchangeability between test and calibration nodes, which are typically satisfied through IID assumption. But in transductive setting, testing nodes are connected to calibration nodes in the graph, thus, it is unclear if they are exchangeable.
1
0
6
Joint work with an all-star team
@marinkazitnik
@TianfanFu
@WenhaoGao1
@yzhao062
!
TDC is a community effort and we are looking for contributors! If interested, fill in this form:
Github:
Website:
1
3
7
With an awesome team:
@TianfanFu
@caoxiao_danica
@jimeng
@marinkazitnik
@AiLucasg
!
Btw, a link to TDC if you want to try out the ADME datasets:
0
1
6
Challenge 1 is that prev models predict diseases with known treatments and rich molecular understanding (Scenario A). But diseases of interest are often neglected, where it has zero/few treatments and limited understanding (e.g. rare diseases, Scenario B). 2/7
1
0
6
7/ This setting directly biases the AL selection because, in low data, the model may yield poor estimates of kernel (relation of unseen perturbations). We confirm this by an independent data analysis⬇️. A bad kernel directly leads to suboptimal selection. How do we solve this?
1
0
6
Had fun with these amazing folks
@LogmlSchool
where we studied if classic network medicine principle (local hypothesis, shared component hypothesis) is automatically encoded in the bio KG embedding! ⬇️
I am excited to share our
@LogmlSchool
Project summary: Exploring network medicine principles encoded by knowledge graphs embeddings.
Thanks to
@KexinHuang5
@KumailAlhamoud
@farhan__tanvir
Aarthi Venkat and Yepeng Huang for the fun time together!
4
11
39
0
2
6
With
@jimeng
@genomestake
@caoxiao_danica
@AiLucasg
Cathy Critchlow
Github repo that hosts links to all papers following the organization in the paper:
0
4
6
Checkout Clinical XLNet () !
We adapt XLNet to clinical context and leverage the sequential dimension of notes.
We did thorough cohort curation with our awesome clinician team
@DanaFarber
and achieve SOTA on Prolonged Mechanical Ventilation prediction!
1
1
6
We conduct a user study with 12 clinicians and show it can improve human accuracy, trust, etc. You can explore it here: 6/7
1
1
5
While ensuring valid coverage, it also achieves strong reduction in prediction set size/interval length compared to vanilla application of conformal prediction to GNN.
1
0
4
A prediction set with enormous size might not be practically desirable even though it achieves valid coverage. We conduct an empirical analysis and find that inefficiencies are correlated among network edges.
1
0
4
This motivates us to develop a topology-aware conformal correction model that adjusts the prediction output. This correction model is learned by simulating the downstream conformal step with a differentiable inefficiency loss.
1
1
4
Come to poster
#60
on Friday at 5:15PM EST
#ML4H
! Paper at
Enjoyable summer work done at
@flatironhealth
, with the amazing
@sankeerthsai23
and
@alex_s_rich
! This work is built upon many great ideas from the ML team
@benbirnbaum
@GriffinAdams16
,….!
0
2
5
Happy to present SkipGNN! We inject skip-similarity into GNN, which is a very powerful property that distincts molecular networks from classic networks! The result is promising on DDI, PPI, GDI, DTI!
Thanks for all the guidance!
@marinkazitnik
@caoxiao_danica
@jimeng
@AiLucasg
Excited to share a new preprint on predicting molecular
#interactions
with skip-graph networks
Skip-
#GNN
uses 2nd-order interactions, which proved incredibly useful in
#bionets
over the last decade
Led by a fantastic student
@KexinHuang5
@harvard_data
0
7
49
0
1
5
9/ How do we incorporate them into the AL selection method? Also, these priors span across various modalities such as text, image, geometrical objects, and matrices. How do we integrate these multi-modalities?
1
0
5
ClinicalBERT is out in () Good performance on intrinsic eval and readmission prediction!
Code and checkpoints:
Greatest logo design by
@thejaan
!
Also, checkout concurrent work by
@Emily_Alsentzer
() !
0
0
4
Challenge 3 is how to use it for clinicians since predictions alone are rarely useful. Using graph XAI, we design a human-centered AI interface where users can explore the molecular mechanisms that drive disease treatments. It enables rapid hypothesis generation. 5/7
1
0
4
I did this in less than a day (!) because of these two great open-sourced softwares! 1. Our DeepPurpose: DL for drug discovery library with < 10 lines of codes: and 2. Gradio: 5 minutes programmatic web UI building: !
1
1
4
If you are an international student (undergrad/master) who wants to be guided by a senior researcher, also consider apply through this link:
2
0
4
CF-GNN also achieved conditional coverage over numerous network features suggesting its robustness!
1
0
3
We develop a theory that enables graph exchangeability and it only requires permutation invariance condition, which is easily satisfied by majority of GNNs. We also obtain an exact characterization of the empirical test-time coverage distribution.
1
0
3
We observe CF-GNN achieves valid marginal coverage while the previous uncertainty quantification methods fail to do so.
1
0
3
Happy to present DeepPurpose: a Deep Learning-based Drug Repurposing Toolkit. Given a target, with ONE line, it gives a ranked list of drugs by aggregating 5 SOTA pretrained models. It finds 3 drugs in current clinical trials for SARS-CoV2 3CLPro! For ML researchers, .. 1/n
1
1
4
Great experience as an intern last summer at Pfizer's ML group, learned a lot from
@MBordyuh
@VishnuSresht
Brajesh Rai
0
1
2
Our work on predicting drug interaction with substructure representation is featured in
@techreview
!
A new artificial intelligence system can better predict adverse drug interactions.
2
28
53
0
0
3
For these diseases, the standard GNN fails mainly because very few nodes connect to them and the embedding is bad. Thus, we propose a method to predict for them by leveraging network medicine principles. We show large improvements over 6 realistic and hard disease splits. 3/7
1
0
3
We open source all the scripts and weights! Including pre-training at . With
@tremblerz
Sitong Chen
@lindvalllab
et al. from the great HST 953 class
@mit_hst
@MITcriticaldata
0
1
3
Come to our workshop tomorrow! Lots of great talks!
Can't wait to see you at 8am-6pm (EST), Monday! Join us in this great event with 7 featured LIVE talks, a panel discussion and 52 contributed papers about
#AI4Science
@NeurIPSConf
Schedule:
Virtual site:
2
17
83
0
0
2
@andrewwhite01
@marinkazitnik
@WangQianwenToo
@payal_chandak
@ngehlenborg
@icmlconf
@HarvardDBMI
here you go:
0
0
3
DeepPurpose provides a 10-lines flexible framework that offers 8 encoders for drugs (Morgan, CNN, MPNN...), 7 for proteins (AAC, CNN, transformers,...), in combination, 50+ models, where most are novel! Switching encoder is as simple as changing the input encoding name! .. 2/n
1
0
2
@A_Aspuru_Guzik
@curiouswavefn
@neurobongo
SELFIES could also mitigate one weakness of the paper (some decomposed substrings of SMILES are not chemically valid, which may sometimes hurt the explainability). Thanks and will give it a try!
0
0
2
@CyrusMaher
@GradioML
Hey Cyrus, glad to hear it is useful! I contacted the Gradio team to get access, I will PM you the contact person’s email address
1
0
2
@david_sontag
@_MiguelHernan
China currently implements a “COVID QR-code” which stratifies each person’s health status into green, yellow, red, based on self-reported info. The lockdown is lifted only on people with green code. Could it serve as a preliminary example?
0
0
2
Lots of additional features! 1. Automatic identification of drug-target interaction (binary) or binding affinity (continuous) prediction given the training dataset, and it will change the metrics and loss function automatically as well .. 4/n
1
0
2
1
0
2
Challenge 2 is model evaluation. In silico data splits could contain confirmatory bias. Thus, we further test in the wild by using external Mount Sinai EHR and show novel predictions made by TxGNN can be validated in off-label prescriptions of millions of patients. 4/7
1
0
2
For one line repurposing, it also accepts customized training datasets such as assay data! It will automatically train five new models and gives out test set performances for each one! We also extend the functionality to the virtual screening! All in one line!! .. 3/n
1
0
1
An example code is also released here: .
0
0
1
1
0
1
@CyrusMaher
Hey Cyrus, we plan to release the project in next Tuesday in will also keep you posted here!
0
0
1
@CyrusMaher
@marinkazitnik
@TianfanFu
@WenhaoGao1
@yzhao062
Hey Cyrus, yes, it is because some source datasets contain this license. So in the first release, we just use the most restricted one. But this is a good point, we will try to find license from individual data source so that industry folks can use them. added to the backlog!
0
0
1
@_romain_lopez_
Thank you for making the internship experience awesome and for all the advice!!
0
0
0
@CyrusMaher
Interesting! I tried to use MLP + Morgan as the drug encoder and it is not as good for DTI, but I indeed found it has comparable and sometimes better performance in many molecular property predictions tasks compared to MPNN and transformers etc. I will try the random forest! Thx!
0
0
1