📎📢ICML WORKSHOP 2025. 1. 2nd Generative AI for Biology Workshop( . 2. Workshop on Multi-modal Foundation Models and Large Language Models for Life Sciences(.

Code:

🚀Fabian Theis (@fabian_theis) paper is here!!. How can we predict gene expression and chromatin accessibility patterns in individual cells directly from DNA sequence, and what insights might this reveal about cellular regulation and disease?@biorxivpreprint @TU_Muenchen

How much data is truly needed to build machine learning models that can reliably predict changes in antibody-antigen binding affinity across diverse protein complexes?@NatComputSci @UniofOxford . "Investigating the volume and diversity of data needed for generalizable

Authors: @jakublala et. al Stefano Angioletti-Uberti.Link:

Can computational protein design be democratized through a flexible, model-agnostic framework that explores energy landscapes to overcome the limitations of current rigid pipelines?@biorxivpreprint @imperialcollege . "BAGEL: Protein Engineering via Exploration of an Energy

Can artificial intelligence predict cancer mutations directly from routine microscopy slides to improve patient care and reduce tissue waste?@NatureMedicine @IcahnMountSinai . "Real-world deployment of a fine-tuned pathology foundation model for lung cancer biomarker detection"

Can AI automation safely reduce radiologist workload in tuberculosis screening while maintaining diagnostic accuracy in high-volume, low-prevalence settings?@npjDigitalMed @M42Health . "Population-scale cross-sectional observational study for AI-powered TB screening on one

Can plasma proteins reveal which organs are aging faster and predict who will live longer by identifying the brain and immune system as key longevity gatekeepers?@NatureMedicine @Stanford . "Plasma proteomics links brain and immune system aging with healthspan and longevity". •

Could the gut microbiome hold the key to predicting which children will succeed or fail with peanut allergy treatment?@NatureComms @UCSF . "Gut microbial bile and amino acid metabolism associate with peanut oral immunotherapy failure". • Peanut oral immunotherapy (POIT) shows

Can we predict and characterize how the human immune system's B cells respond to a two-dose Ebola vaccine regimen?@NatureComms @UniofOxford . "Prediction and characterisation of the human B cell response to a heterologous two-dose Ebola vaccine". • Ebola virus disease outbreaks

How can we mathematically decode the hidden rules that transform brain anatomy into functional networks?@arxiv @UnivOfKansas . "A Linear Generative Framework for Structure-Function Coupling in the Human Brain". • Brain function emerges from coordinated activity across

Can wearable devices predict surgical complications in children before doctors even know they exist?@ScienceAdvances . "Biorhythms derived from consumer wearables predictpostoperative complications in children". • Approximately 14% of children develop postoperative

What if protein-protein interaction prediction models could actually understand the physical mechanisms underlying interactions rather than just memorizing sequence patterns?@biorxivpreprint . "Mechanism-Aware Protein-Protein Interaction Prediction via Contact-Guided Dual

How can scientists efficiently map and predict regulatory DNA sequences across different vertebrate species to understand the fundamental rules governing gene expression? @CellCellPress . "Modeling the vertebrate regulatory sequence landscape by UUATAC-seq and deep learning". •

How can federated learning enable fair evaluation of healthcare AI algorithms across multiple international institutions while preserving data privacy?@NatureComms @HeidelbergU @IndianaUniv . "Towards fair decentralized benchmarking of healthcare AI algorithms with the Federated

Can machine learning models effectively diagnose autoimmune diseases from T cell receptor repertoires despite the challenges of sequence sparsity and low witness rates?@arxiv @Tsinghua_Uni . "Classification of autoimmune diseases from Peripheral blood TCR repertoires by

Can a unified multimodal model effectively integrate cellular morphology, gene expression, and spatial organization across biological scales to enable both predictive and generative tasks in spatial transcriptomics?@arxiv @Harvard . "SPATIA: Multimodal Model for Prediction and

Can machine learning models accurately replace computationally expensive finite element simulations to predict bone fracture healing trajectories in real-time?@sciencedirect @harvardmed . "Mechanobiology-guided machine learning models for predicting long bone fracture healing

Could gut bacteria help predict which colorectal cancer patients will develop lymph node metastasis before surgery?@ASMicrobiology . "Identification and predictive machine learning models construction of gut microbiota associated with lymph node metastasis in colorectal cancer"