aixbiobot.bsky.social
AI x Bio Discovery
@aixbiobot.bsky.social
610 followers 1 following 4.3K posts
Automated discovery of AI x Bio preprint papers.
Posts Media Videos Starter Packs
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 4h
DeepEmbCas9: Cas9 coevolution and sgRNA structural information for CRISPR-Cas9 cleavage activity prediction [new]
DeepEmbCas9: Models Cas9 cleavage w/ embeddings, shows key domains affect activity.
DeepEmbCas9: Cas9 coevolution and sgRNA structural information for CRISPR-Cas9 cleavage activity prediction Figure 1 Figure 2 Figure 3
1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 5h
JAAG: a JSON input file Assembler for AlphaFold 3 with Glycan integration [new]
Facilitates AlphaFold 3 modeling of biomolecules, including glycans, by simplifying JSON input file creation with a web-based interface.
Figure 1 Figure 2
1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 5h
Unified Sampling and Ranking for Protein Docking with DFMDock [updated]
Protein Docking: Unified Sampling/Ranking uses diffusion, predicting denoising forces/energy for decoy ranking.
Unified Sampling and Ranking for Protein Docking with DFMDock Figure 1 Figure 3 Figure 2
1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 6h
UniOTalign: A Global Matching Framework for Protein Alignment via Optimal Transport [new]
Global matching via Optimal Transport allows alignment by finding an optimal flow between proteins represented as residue distributions.
UniOTalign: A Global Matching Framework for Protein Alignment via Optimal Transport Table 1
1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 6h
Evolutionary Profiles for Protein Fitness Prediction [new]
Leveraging evolution as reward maximization, integrates within-family profiles and cross-family structural constraints to predict mutational fitness.
Evolutionary Profiles for Protein Fitness Prediction Figure 1 Table 1 Figure 2
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 6h
Uncovering distinct protein conformations using coevolutionary information and AlphaFold [new]
MSA-guided sampling of protein folds via residue frequencies & coevolution.
Uncovering distinct protein conformations using coevolutionary information and AlphaFold Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 7h
Optimal Gene Panel Selection for Targeted Spatial Transcriptomics Experiments [new]
Optimizing gene panel selection...
...for spatial transcriptomics is achieved via a novel method using scRNA-seq data and a gated autoencoder.
Optimal Gene Panel Selection for Targeted Spatial Transcriptomics Experiments Figure 1 Figure 2
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 8h
The functional landscape of the human ubiquitinome [new]
Maps human ubiquitination sites, identifies conserved sites likely regulating signaling, and scores sites to predict functional roles validated via experiments.
The functional landscape of the human ubiquitinome Figure 1 Figure 2 Figure 3
1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 10h
Nanobody-Driven Stabilization Synergistically Rescues F508del-CFTR and Reveals an Alternative Active State of the Channel [new]
Nanobody stabilizes CFTR, boosting channel activity & revealing active conformation in cystic fibrosis.
Nanobody-Driven Stabilization Synergistically Rescues F508del-CFTR and Reveals an Alternative Active State of the Channel Figure S1 Figure 1 Figure S2
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 10h
Finding the perfect promoter: Correlating single-cell transcriptome data with gene drive performance [new]
scRNA/gene drive correlation reveals spatially/temporal Cas9 promoter optimum.
Finding the perfect promoter: Correlating single-cell transcriptome data with gene drive performance Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 10h
Sequence-free landscape inference for directed evolution [new]
Infers landscape ruggedness in directed evolution using population phenotype & mutation rate data. This informs selection strategies for improved outcomes.
Sequence-free landscape inference for directed evolution Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 11h
From Genomics Alterations to Expression Dynamics: A Hierarchical Multi-Omics Integration Framework with gINTomics [new]
Multi-omics link genomic changes to transcriptional effects, finding regulatory mechanisms & biomarkers.
From Genomics Alterations to Expression Dynamics: A Hierarchical Multi-Omics Integration Framework with gINTomics Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 12h
NUCGEN3D: A synthetic framework for large-scale 3D nuclear segmentation with open-source training data and models [new]
NUCGEN3D: Synthetic 3D microscopy data framework aids robust 3D nuclear segmentation, solving data lack.
NUCGEN3D: A synthetic framework for large-scale 3D nuclear segmentation with open-source training data and models Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 13h
Improving the Accuracy of Distance-Based Protein-Ligand Binding Affinity Prediction Using Linear Regression and Artificial Neural Network [new]
Improving the Accuracy of Distance-Based Protein-Ligand Binding Affinity Prediction Using Linear Regression and Artificial Neural Network Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 17h
Uncovering the Mechanistic Landscape of Regulatory DNA with Deep Learning [new]
...maps how mutations alter transcription factor binding at regulatory loci, revealing robust and reprogrammable signals.
Uncovering the Mechanistic Landscape of Regulatory DNA with Deep Learning Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 22h
Localization of Macromolecules in Crowded Cellular Cryo-electron Tomograms from Extremely Sparse Labels [updated]
Macromolecule cryo-ET localization via annot.-efficient pos.-unlabeled learning.
Localization of Macromolecules in Crowded Cellular Cryo-electron Tomograms from Extremely Sparse Labels Figure 1 Figure 2 Figure 4
1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 22h
Homomorphic Encryption: An Application to Polygenic Risk Scores [updated]
Applies homomorphic encryption to polygenic risk scores, enabling privacy-preserving calculations via a three-party system using the CKKS protocol.
Homomorphic Encryption: An Application to Polygenic Risk Scores Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 22h
mosna reveals different types of cellular interactions predictive of response to immunotherapies and survival in cancer [updated]
Spatial omics & clinical data: cell interaction patterns predict immunotherapy response & survival.
Figure 2 Figure 3 Figure 4 Figure 5
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 22h
Dominant-negative TP53 mutations potentiated by the HSF1-regulatedproteostasis network [updated]
HSF1 activation favors dominant-negative p53 mutations, esp. in the DNA-binding domain, shaping oncogenic mutations.
Dominant-negative TP53 mutations potentiated by the HSF1-regulatedproteostasis network Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 1d
Accelerating scRNA-seq analysis: automated cell type annotation using representation learning and vector search [new]
Automates cell annotation via expression profiles vs. ref; provides annotations w/o pre-defined markers.
Accelerating scRNA-seq analysis: automated cell type annotation using representation learning and vector search Figure 1 Figure 2 Table 2
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 1d
GOFlowLLM - Curating miRNA literature with Large Language Models and flowcharts [new]
Uses LLMs and GO flowcharts to automate miRNA curation, extracting gene silencing data for GO annotation at scale and with reasoning traces.
GOFlowLLM - Curating miRNA literature with Large Language Models and flowcharts Figure 1 Figure 2 Figure 3
1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 1d
SpaDiff: Denoising for Sequence-based Spatial Transcriptomics via Diffusion Process [new]
RNA displacement simulated, reversed: restores spatial data & count integrity.
SpaDiff: Denoising for Sequence-based Spatial Transcriptomics via Diffusion Process Figure 1 Figure 2 Figure 3
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 1d
Citizen science gamers enable automated flow cytometry gating through machine learning [new]
Crowdsourced cytometry annotations train ML for automated, reproducible gating, aiding immunology.
Figure 1 Figure 2 Figure 3 Figure 4
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 1d
Artificial Intelligence-enabled Histological Analysis in Preclinical Respiratory Disease Models: A Scoping Review [new]
AI histology analysis in resp. models shows variable validation, reporting. Standard workflows req'd.
Figure 1 Figure 2 Figure 3 Table 1
aixbiobot.bsky.social
AI x Bio Discovery @aixbiobot.bsky.social · 1d
Protein Language Models are Accidental Taxonomists [new]
Exploiting taxonomic biases in PPI data, where models learn species origin instead of true interaction features. Removing this bias reduces performance.
Protein Language Models are Accidental Taxonomists Figure 1 Figure 3 Figure 2
2