diff --git a/feed.xml b/feed.xml
index 6a665e01..6d9ed20e 100644
--- a/feed.xml
+++ b/feed.xml
@@ -1 +1 @@
-Jekyll2024-02-17T14:12:55-05:00https://zitniklab.hms.harvard.edu/feed.xmlZitnik LabHarvard Machine Learning for Medicine and ScienceMarinka ZitnikKaneb Fellowship and Dean’s Innovation Award2024-02-02T00:00:00-05:002024-02-02T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/02/02/KanebFellowship<p>The lab receives the <a href="#">John and Virginia Kaneb Fellowship Award at Harvard Medical School</a>. The lab also receives the <a href="#">Dean’s Innovation Award for the Use of Artificial Intelligence.</a></p>Marinka ZitnikThe lab receives the John and Virginia Kaneb Fellowship Award at Harvard Medical School. The lab also receives the Dean’s Innovation Award for the Use of Artificial Intelligence.NSF CAREER Award2024-02-02T00:00:00-05:002024-02-02T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/02/02/NSFCAREER<p>The lab receives the <a href="#">NSF CAREER Award</a> for our research in geometric deep learning to facilitate algorithmic and scientific advances in therapeutics.</p>Marinka ZitnikThe lab receives the NSF CAREER Award for our research in geometric deep learning to facilitate algorithmic and scientific advances in therapeutics.AI’s Prospects in Nature Machine Intelligence2024-01-25T00:00:00-05:002024-01-25T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/01/25/MultimodalLearning<p>We discussed <a href="https://www.nature.com/articles/s42256-023-00784-5">AI’s 2024 prospects with Nature Machine Intelligence</a>, covering LLM progress, multimodal AI, multi-task agents, and how to bridge the digital divide across communities and world regions.</p>Marinka ZitnikWe discussed AI’s 2024 prospects with Nature Machine Intelligence, covering LLM progress, multimodal AI, multi-task agents, and how to bridge the digital divide across communities and world regions.Combinatorial Therapeutic Perturbations2024-01-01T00:00:00-05:002024-01-01T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/01/01/PDGrapher<p>New paper introducing <a href="https://www.biorxiv.org/content/10.1101/2024.01.03.573985">PDGrapher for combinatorial prediction of chemical and genetic perturbations</a> using causally-inspired neural networks.</p>Marinka ZitnikNew paper introducing PDGrapher for combinatorial prediction of chemical and genetic perturbations using causally-inspired neural networks.Next Generation of Therapeutics Commons2023-11-23T00:00:00-05:002023-11-23T00:00:00-05:00https://zitniklab.hms.harvard.edu/2023/11/23/OpenPositions<p>We are building the next generation of <a href="https://tdcommons.ai/">Therapeutics Commons</a>! We are seeking <a href="/postdoc-TDC/">outstanding fellows who will lead AI research to advance molecular drug design and clinical drug development.</a></p>Marinka ZitnikWe are building the next generation of Therapeutics Commons! We are seeking outstanding fellows who will lead AI research to advance molecular drug design and clinical drug development.Graph AI in Medicine2023-10-24T00:00:00-04:002023-10-24T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/10/24/GraphAIMedicine<p><a href="https://arxiv.org/abs/2310.13767">Graph AI models in medicine</a> integrate diverse data modalities through pre-training, facilitate interactive feedback loops, and foster human-AI collaboration, paving the way to clinically meaningful predictions.</p>Marinka ZitnikGraph AI models in medicine integrate diverse data modalities through pre-training, facilitate interactive feedback loops, and foster human-AI collaboration, paving the way to clinically meaningful predictions.Structure-Based Drug Design2023-10-24T00:00:00-04:002023-10-24T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/10/24/SBDD<p><a href="https://arxiv.org/abs/2306.11768">Geometric deep learning has emerged as a valuable tool</a> for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.</p>Marinka ZitnikGeometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.New papers accepted at NeurIPS2023-09-22T00:00:00-04:002023-09-22T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/09/22/NewPapersNeurIPS<p>Congratulations to <a href="https://zitniklab.hms.harvard.edu/people/">Owen and Zaixi</a> for having their papers accepted <a href="https://nips.cc/">as spotlights at NeurIPS</a>! These papers introduce techniques for <a href="https://arxiv.org/abs/2306.02109">explaining time series models through self-supervised learning</a> and <a href="https://arxiv.org/abs/2310.02553">co-designing protein pocket sequences & 3D structures</a>.</p>Marinka ZitnikCongratulations to Owen and Zaixi for having their papers accepted as spotlights at NeurIPS! These papers introduce techniques for explaining time series models through self-supervised learning and co-designing protein pocket sequences & 3D structures.Future Directions in Network Biology2023-09-17T00:00:00-04:002023-09-17T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/09/17/Future-NetBio<p>Excited to share our perspectives on <a href="https://arxiv.org/abs/2309.08478">current and future directions in network biology.</a></p>Marinka ZitnikExcited to share our perspectives on current and future directions in network biology.Scientific Discovery in the Age of AI2023-08-02T00:00:00-04:002023-08-02T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/08/02/AI4Science-Nature<p>New paper on the role of <a href="https://rdcu.be/dinBA">artificial intelligence in scientific discovery is published in Nature.</a></p>Marinka ZitnikNew paper on the role of artificial intelligence in scientific discovery is published in Nature.
\ No newline at end of file
+Jekyll2024-03-02T01:04:44-05:00https://zitniklab.hms.harvard.edu/feed.xmlZitnik LabHarvard Machine Learning for Medicine and ScienceMarinka ZitnikPocketGen - Generating Full-Atom Ligand-Binding Protein Pockets2024-03-01T00:00:00-05:002024-03-01T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/03/01/PocketGen<p><a href="https://www.biorxiv.org/content/10.1101/2024.02.25.581968">PocketGen is a deep generative model</a> that generates residue sequence and full-atom structure of protein pockets, maximizing binding to ligands. <a href="https://zitniklab.hms.harvard.edu/projects/PocketGen/">Project website.</a></p>Marinka ZitnikPocketGen is a deep generative model that generates residue sequence and full-atom structure of protein pockets, maximizing binding to ligands. Project website.SPECTRA - Generalizability of Molecular AI2024-02-28T00:00:00-05:002024-02-28T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/02/28/SPECTRA-model-generalizability<p><a href="https://www.biorxiv.org/content/10.1101/2024.02.25.581982v1">SPECTRA is an approach for holistic evaluation of how AI models</a> generalize to new molecular datasets. <a href="https://zitniklab.hms.harvard.edu/projects/SPECTRA/">Project website.</a></p>Marinka ZitnikSPECTRA is an approach for holistic evaluation of how AI models generalize to new molecular datasets. Project website.Kaneb Fellowship and Dean’s Innovation Award2024-02-02T00:00:00-05:002024-02-02T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/02/02/KanebFellowship<p>The lab receives the <a href="#">John and Virginia Kaneb Fellowship Award at Harvard Medical School</a>. The lab also receives the <a href="#">Dean’s Innovation Award for the Use of Artificial Intelligence.</a></p>Marinka ZitnikThe lab receives the John and Virginia Kaneb Fellowship Award at Harvard Medical School. The lab also receives the Dean’s Innovation Award for the Use of Artificial Intelligence.NSF CAREER Award2024-02-02T00:00:00-05:002024-02-02T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/02/02/NSFCAREER<p>The lab receives the <a href="#">NSF CAREER Award</a> for our research in geometric deep learning to facilitate algorithmic and scientific advances in therapeutics.</p>Marinka ZitnikThe lab receives the NSF CAREER Award for our research in geometric deep learning to facilitate algorithmic and scientific advances in therapeutics.AI’s Prospects in Nature Machine Intelligence2024-01-25T00:00:00-05:002024-01-25T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/01/25/MultimodalLearning<p>We discussed <a href="https://www.nature.com/articles/s42256-023-00784-5">AI’s 2024 prospects with Nature Machine Intelligence</a>, covering LLM progress, multimodal AI, multi-task agents, and how to bridge the digital divide across communities and world regions.</p>Marinka ZitnikWe discussed AI’s 2024 prospects with Nature Machine Intelligence, covering LLM progress, multimodal AI, multi-task agents, and how to bridge the digital divide across communities and world regions.Combinatorial Therapeutic Perturbations2024-01-01T00:00:00-05:002024-01-01T00:00:00-05:00https://zitniklab.hms.harvard.edu/2024/01/01/PDGrapher<p>New paper introducing <a href="https://www.biorxiv.org/content/10.1101/2024.01.03.573985">PDGrapher for combinatorial prediction of chemical and genetic perturbations</a> using causally-inspired neural networks.</p>Marinka ZitnikNew paper introducing PDGrapher for combinatorial prediction of chemical and genetic perturbations using causally-inspired neural networks.Next Generation of Therapeutics Commons2023-11-23T00:00:00-05:002023-11-23T00:00:00-05:00https://zitniklab.hms.harvard.edu/2023/11/23/OpenPositions<p>We are building the next generation of <a href="https://tdcommons.ai/">Therapeutics Commons</a>! We are seeking <a href="/postdoc-TDC/">outstanding fellows who will lead AI research to advance molecular drug design and clinical drug development.</a></p>Marinka ZitnikWe are building the next generation of Therapeutics Commons! We are seeking outstanding fellows who will lead AI research to advance molecular drug design and clinical drug development.Graph AI in Medicine2023-10-24T00:00:00-04:002023-10-24T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/10/24/GraphAIMedicine<p><a href="https://arxiv.org/abs/2310.13767">Graph AI models in medicine</a> integrate diverse data modalities through pre-training, facilitate interactive feedback loops, and foster human-AI collaboration, paving the way to clinically meaningful predictions.</p>Marinka ZitnikGraph AI models in medicine integrate diverse data modalities through pre-training, facilitate interactive feedback loops, and foster human-AI collaboration, paving the way to clinically meaningful predictions.Structure-Based Drug Design2023-10-24T00:00:00-04:002023-10-24T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/10/24/SBDD<p><a href="https://arxiv.org/abs/2306.11768">Geometric deep learning has emerged as a valuable tool</a> for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.</p>Marinka ZitnikGeometric deep learning has emerged as a valuable tool for structure-based drug design, to generate and refine biomolecules by leveraging detailed three-dimensional geometric and molecular interaction information.New papers accepted at NeurIPS2023-09-22T00:00:00-04:002023-09-22T00:00:00-04:00https://zitniklab.hms.harvard.edu/2023/09/22/NewPapersNeurIPS<p>Congratulations to <a href="https://zitniklab.hms.harvard.edu/people/">Owen and Zaixi</a> for having their papers accepted <a href="https://nips.cc/">as spotlights at NeurIPS</a>! These papers introduce techniques for <a href="https://arxiv.org/abs/2306.02109">explaining time series models through self-supervised learning</a> and <a href="https://arxiv.org/abs/2310.02553">co-designing protein pocket sequences & 3D structures</a>.</p>Marinka ZitnikCongratulations to Owen and Zaixi for having their papers accepted as spotlights at NeurIPS! These papers introduce techniques for explaining time series models through self-supervised learning and co-designing protein pocket sequences & 3D structures.
\ No newline at end of file
diff --git a/img/PocketGen-1.png b/img/PocketGen-1.png
new file mode 100644
index 00000000..fdd12591
Binary files /dev/null and b/img/PocketGen-1.png differ
diff --git a/img/PocketGen-2.png b/img/PocketGen-2.png
new file mode 100644
index 00000000..9a6f1377
Binary files /dev/null and b/img/PocketGen-2.png differ
diff --git a/img/PocketGen-3.png b/img/PocketGen-3.png
new file mode 100644
index 00000000..2d8fc9bb
Binary files /dev/null and b/img/PocketGen-3.png differ
diff --git a/img/PocketGen-4.png b/img/PocketGen-4.png
new file mode 100644
index 00000000..6a27ef5e
Binary files /dev/null and b/img/PocketGen-4.png differ
diff --git a/img/kushan_weerakoon.png b/img/kushan_weerakoon.png
new file mode 100644
index 00000000..9e62cffa
Binary files /dev/null and b/img/kushan_weerakoon.png differ
diff --git a/index.html b/index.html
index 835da1e9..94d5da1e 100644
--- a/index.html
+++ b/index.html
@@ -177,6 +177,62 @@
AI for Science | Therapeutic Science
+
+
+
+
+
+
Mar 2024: PocketGen - Generating Full-Atom Ligand-Binding Protein Pockets
The MITxHarvard Women in AI initiative talked with Marinka about AI, machine learning, and the role of new technologies in biomedical research.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Aug 2020: Trustworthy AI for Healthcare
-
-
-
-
-
-
-
-
-
We are excited to be co-organizing a workshop at AAAI 2021 on Trustworthy AI for Healthcare! We have a stellar lineup of speakers. Details to follow soon!
-
-
-
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
Sep 2020: MITxHarvard Women in AI Interview
+
+
+
+
+
+
+
+
+
The MITxHarvard Women in AI initiative talked with Marinka about AI, machine learning, and the role of new technologies in biomedical research.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Aug 2020: Trustworthy AI for Healthcare
+
+
+
+
+
+
+
+
+
We are excited to be co-organizing a workshop at AAAI 2021 on Trustworthy AI for Healthcare! We have a stellar lineup of speakers. Details to follow soon!
diff --git a/products/ada_fang/index.html b/products/ada_fang/index.html
index a2136c6e..fcd99bed 100644
--- a/products/ada_fang/index.html
+++ b/products/ada_fang/index.html
@@ -23,14 +23,14 @@
-
+
+{"image":"https://zitniklab.hms.harvard.edu/img/ada_fang.png","url":"https://zitniklab.hms.harvard.edu/products/ada_fang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/ada_fang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Ada Fang","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/alex_verce.png","url":"https://zitniklab.hms.harvard.edu/products/alejandro_velez_arce/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/alejandro_velez_arce/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Alejandro Velez Arce","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/andrew_shen.png","url":"https://zitniklab.hms.harvard.edu/products/andrew_shen/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/andrew_shen/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Andrew Shen","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/ayush_noori.png","url":"https://zitniklab.hms.harvard.edu/products/ayush_noori/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/ayush_noori/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Ayush Noori","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/george_dasoulas.png","url":"https://zitniklab.hms.harvard.edu/products/george_dasoulas/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/george_dasoulas/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"George Dasoulas","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/ivy_liang.png","url":"https://zitniklab.hms.harvard.edu/products/ivy_liang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/ivy_liang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Ivy Liang","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/jason_poulos.png","url":"https://zitniklab.hms.harvard.edu/products/jason_poulos/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/jason_poulos/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Jason Poulos","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/jonathan_schwarz.png","url":"https://zitniklab.hms.harvard.edu/products/jonathan_schwarz/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/jonathan_schwarz/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Jonathan Richard Schwarz","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
-Josh Pan | Zitnik Lab
+Kushan Weerakoon | Zitnik Lab
-
+
-
-
+
+
-
+
-
+
-
-
+
+
+{"image":"https://zitniklab.hms.harvard.edu/img/kushan_weerakoon.png","url":"https://zitniklab.hms.harvard.edu/products/kushan_weerakoon/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/kushan_weerakoon/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Kushan Weerakoon","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/lei_huang.png","url":"https://zitniklab.hms.harvard.edu/products/lei_huang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/lei_huang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Lei Huang","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/marinka_zitnik.png","url":"https://zitniklab.hms.harvard.edu/products/marinka_zitnik/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/marinka_zitnik/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Marinka Zitnik","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/michelle_dai.png","url":"https://zitniklab.hms.harvard.edu/products/michelle_dai/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/michelle_dai/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Michelle Dai","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/michelle_li.png","url":"https://zitniklab.hms.harvard.edu/products/michelle_li/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/michelle_li/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Michelle M. Li","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/owen_queen.png","url":"https://zitniklab.hms.harvard.edu/products/owen_queen/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/owen_queen/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Owen Queen","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/ruth_johnson.png","url":"https://zitniklab.hms.harvard.edu/products/ruth_johnson/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/ruth_johnson/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Ruth Johnson","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/shanghua_gao.png","url":"https://zitniklab.hms.harvard.edu/products/shanghua_gao/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/shanghua_gao/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Shanghua Gao","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/tianlong_chen.png","url":"https://zitniklab.hms.harvard.edu/products/tianlong_chen/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/tianlong_chen/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Tianlong Chen","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/valentina_giunchiglia.png","url":"https://zitniklab.hms.harvard.edu/products/valentina_giunchiglia/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/valentina_giunchiglia/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Valentina Giunchiglia","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/varun_ullanat.png","url":"https://zitniklab.hms.harvard.edu/products/varun_ullanat/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/varun_ullanat/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Varun Ullanat","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/wanxiang_shen.png","url":"https://zitniklab.hms.harvard.edu/products/wanxiang_shen/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/wanxiang_shen/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Wanxiang Shen","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/xiang_lin.png","url":"https://zitniklab.hms.harvard.edu/products/xiang_lin/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/xiang_lin/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Xiang Lin","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/xiaorui_su.png","url":"https://zitniklab.hms.harvard.edu/products/xiaorui_su/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/xiaorui_su/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Xiaorui Su","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/yasha_ektefaie.png","url":"https://zitniklab.hms.harvard.edu/products/yasha_ektefaie/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/yasha_ektefaie/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Yasha Ektefaie","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Biomedical Machine Learning, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/yepeng_huang.png","url":"https://zitniklab.hms.harvard.edu/products/yepeng_huang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/yepeng_huang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Yepeng Huang","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+{"image":"https://zitniklab.hms.harvard.edu/img/zaixi_zhang.png","url":"https://zitniklab.hms.harvard.edu/products/zaixi_zhang/","mainEntityOfPage":{"@type":"WebPage","@id":"https://zitniklab.hms.harvard.edu/products/zaixi_zhang/"},"author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Zaixi Zhang","dateModified":"2024-03-02T01:04:44-05:00","description":"Artificial Intelligence (AI), Medicine, Science, and Drug Discovery","datePublished":"2024-03-02T01:04:44-05:00","@type":"BlogPosting","@context":"https://schema.org"}
+
+PocketGen : Generating Full-Atom Ligand-Binding Protein Pockets | Zitnik Lab
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
PocketGen : Generating Full-Atom Ligand-Binding Protein Pockets
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Designing small-molecule-binding proteins, such as enzymes and biosensors, is crucial in protein biology and bioengineering. Generating the protein pocket—the area where the protein interacts with the ligand molecule—presents challenges due to the interactions between ligand molecules and proteins, the flexibility of ligand molecules, the chemistry of amino acid side chains, and multimodal sequence-structure dependencies.
+
+
+
+We introduce PocketGen, a deep generative method for generating the residue sequence and the full-atom structure within the protein pocket region that leverages sequence-structure consistency. PocketGen consists of a bilevel graph transformer for structural encoding and a sequence refinement module that uses a protein language model (pLM) for sequence prediction. The bilevel graph transformer captures interactions at multiple granularities (atom-level and residue/ligand-level) and aspects (intra-protein and protein-ligand) with bilevel attention mechanisms.
+
+
+
+Experiments and case studies indicate that PocketGen can efficiently generate protein pockets with higher binding affinity and validity than state-of-the-art methods. PocketGen is 10 times faster than physics-based methods and achieves a 95% success rate (the ratio of cases that generated pockets has higher binding affinity than the reference ones) with over 64% amino acid recovery rate on the Binding MOAD dataset.
+
+
+
+
Designing Full-Atom Ligand-Binding Protein Pockets
+
+
A primary method for modulating protein functions involves the interaction between proteins and small molecule ligands. These interactions play a critical role in biological processes, such as enzymatic catalysis, signal transduction, and regulatory mechanisms within cells. The binding of small molecules to specific sites on proteins can induce conformational changes, modulate activity, or inhibit functions. This mechanism serves as a valuable tool for studying protein functions and designing small molecule-binding proteins with customized properties for therapeutic and industrial applications, including designing enzymes to catalyze reactions that do not have natural catalysts and developing biosensors that can detect compounds in the environment by transducing signals which can be used for environmental monitoring, clinical diagnostics, pathogen detection, drug delivery systems, and applications in food industry.
+
+
These designs often involve modifying existing ligand-binding protein pockets to facilitate more precise interactions with specific ligands. The complexity of ligand molecule-protein interactions, ligand and side chain flexibility, and sequence-structure relationships, however, pose significant challenges for computational generation of high-validity, ligand-binding protein pockets.
+
+
Overview of PocketGen
+
+
PocketGen implements a co-design scheme, where the model concurrently predicts the sequence and structure of the protein pocket based on the ligand molecule and the protein scaffold (excluding the pocket). PocketGen comprises two modules: the bilevel graph transformer and the sequence refinement module.
+
+
+
+
+
+
To achieve end-to-end molecular generation, PocketGen models the protein-ligand complex as a geometric graph of blocks to handle variable atom counts across different residues and ligands. Initially, the pocket residues are assigned the maximum possible number of atoms (14 atoms) and later are mapped back to specific residue types post-generation. The bilevel graph transformer captures interactions at multiple granularities (atom-level and residue/ligand-level) and aspects (intra-protein and protein-ligand) using bilevel attention mechanisms. To account for the influence of the redesigned pocket on the ligand, the ligand structure is updated during refinement to reflect potential changes in binding pose.
+
+
+
+
+
+
To ensure consistency between protein sequence and structure domains and incorporate evolutionary information encoded in protein language models (the ESM series models), PocketGen implements a structural adapter into protein sequence updates. This adapter facilitates cross-attention between sequence and structure features, promoting information flow and achieving sequence-structure consistency. During training, only the adapter is fine-tuned, while the remaining layers of the protein language models remain unchanged.
+
+
+
+
+
+
Evaluations of PocketGen
+
+
We evaluated PocketGen on redesigning pockets of ligand-binding antibodies, enzymes, and biosensors for the following target ligand molecules:
+
+
+
Antibodies: Cortisol (HCY) is a primary stress hormone that elevates glucose levels in the bloodstream and serves as a biomarker for stress and other conditions. We redesign the pocket of a cortisol-specific antibody (PDB ID 8cby) to potentially aid in the development of immunoassays.
+
Enzymes: Apixaban (APX) is an oral anticoagulant approved by the FDA in 2012 for patients with non-valvular atrial fibrillation to reduce the risk of stroke and blood clots. Apixaban targets Factor Xa (fXa) (PDB ID 2p16), a crucial enzyme in blood coagulation that transforms prothrombin into thrombin for clot formation. Redesigning the pocket of fXa could therefore have therapeutic significance.
+
Biosensors: Fentanyl (7V7) has become a widely abused drug contributing to the opioid crisis. Computational design of fentanyl-binding proteins (biosensors) can facilitate detection and neutralization of the toxin. For example, Baker et al. developed a biosensor (PDB ID 5tzo) for detecting fentanyl in plants.
+
+
+
PLIP is employed to describe the interactions between the designed protein pocket and ligands, comparing these predicted interactions to the original patterns. The pockets produced by PocketGen replicate most non-bonded interactions observed in experimentally measured protein-ligand complexes, and introduce additional physically plausible interaction patterns not present in the original complexes.
Graph AI models in medicine integrate diverse data modalities through pre-training, facilitate interactive feedback loops, and foster human-AI collaboration, paving the way to clinically meaningful predictions.
Apr 2023: Celebrating Achievements of Our Undergrads
+
+
+
+
+
+
+
+
+
Undergraduate researchers Ziyuan, Nick, Yepeng, Jiali, Julia, and Marissa are moving onto their PhD research in Computer Science, Systems Biology, Neuroscience, and Biological & Medical Sciences at Harvard, MIT, Carnegie Mellon University, and UMass Lowell. We are excited for the bright future they created for themselves.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Apr 2023: Welcoming a New Postdoctoral Fellow
+
+
+
+
+
+
+
+
+
An enthusiastic welcome to Tianlong Chen, our newly appointed postdoctoral fellow.
diff --git a/projects/SPECTRA/index.html b/projects/SPECTRA/index.html
index e461a1a8..30754e5f 100644
--- a/projects/SPECTRA/index.html
+++ b/projects/SPECTRA/index.html
@@ -4,16 +4,16 @@
- Evaluating Generalizability of Artificial Intelligence for Molecular Data - Zitnik Lab
+ Evaluating Generalizability of Artificial Intelligence Models for Molecular Datasets - Zitnik Lab
-Evaluating Generalizability of Artificial Intelligence for Molecular Data | Zitnik Lab
+Evaluating Generalizability of Artificial Intelligence Models for Molecular Datasets | Zitnik Lab
-
+
@@ -22,11 +22,11 @@
-
+
+{"url":"https://zitniklab.hms.harvard.edu/projects/SPECTRA/","author":{"@type":"Person","name":"Marinka Zitnik"},"headline":"Evaluating Generalizability of Artificial Intelligence Models for Molecular Datasets","description":"SPECTRA paves the way for a more comprehensive evaluation of foundation models in molecular biology.","@type":"WebPage","@context":"https://schema.org"}
Xiang Lin" />
+ 
Jonathan Richard Schwarz" />
Jason Poulos" />
+ 
George Dasoulas" />
Ruth Johnson" />
+ 
Shanghua Gao" />
+ 
Wanxiang Shen" />
Tianlong Chen" />
+ 
Lei Huang" />
+ 
Xiaorui Su" />
Kushan Weerakoon" />
+ 
Xiang Zhang" />
- 
Chirag Agarwal" />
- 
Josh Pan" />
- 
+
+
+
+