TrendBio

BACKFILLED Code and industry signals are partial for backfilled days.

Rank 1

Single-cell

Learned models reconstruct cell-state dynamics and tissue context from single-cell data

Scalable neural and optimal-transport models now address atlas integration, spatial domains and trajectory reconstruction from single-cell and spatial data.

7 preprints · 8 published · ↑ 3.0× vs 4-wk avg
Rank 2

ML / AI methods

Deep learning fuses imaging, connectivity, methylation and molecular data for brain…

Learned models now integrate cortical geometry, dynamic connectivity, DNA methylation and compound profiles for neurological prediction tasks.

3 preprints · 2 published · ↑ 2.0× vs 4-wk avg
Rank 3

ML / AI methods

Attention-based models target peptide, protein, and drug-target binding prediction

Pretrained encoders and interpretable attention now predict peptide permeability, binding affinity, and protein function from sequence, graph, and structure.

3 preprints · 6 published · ↑ 1.8× vs 4-wk avg
Rank 4

Tools & infrastructure

Modular software packages wrap multi-step omics analyses into reusable, orchestrated…

Recent tools package spatial, single-cell, long-read and multi-omics analyses as open, interface-driven software for reuse.

1 preprint · 10 published · ↑ 1.9× vs 4-wk avg
Rank 5

Genomics

Chromosome-scale and graph pan-genomes resolve lineage-specific genome architecture

Recent assemblies from date palm, cotton, bee, and 90 arachnid genomes expand lineage-resolved comparisons of genome evolution.

1 preprint · 4 published · ↑ 1.7× vs 4-wk avg
Rank 6 Domain-aware priors and cross-modal representations improve biomedical machine learning↑ 1.7×ML / AI methodsRank 7 Functional and ancestry-aware models refine genetic architecture inference↑ 1.7×GenomicsRank 8 Learned models extend RNA folding, tRNA nanopore decoding, and protein alignment↑ 1.7×Structural bio & protein designRank 9 Epigenetic state constrains regulatory dynamics and chromatin mechanics in theory↑ 1.8×EpigenomicsRank 10 Generative protein design adds preference and physics objectives to structure models↑ 1.5×Structural bio & protein designRank 11 Computational methods infer spatial and single-cell structure from transcriptomic dataSingle-cellRank 12 Statistical and classifier methods for omics feature selection and inferenceML / AI methodsRank 13 Software advances haplotype-resolved assembly and repeat-aware genome analysisGenomicsRank 14 Learned and ensemble models extend protein structure and interaction predictionTools & infrastructureRank 15 Protein AI Models Benchmarked Across Antibody Developability and Design TasksML / AI methodsRank 16 Protein language models and simulations link sequence to condition-dependent biophysicsStructural bio & protein designRank 17 Biological analysis tools are coupling software to curated data and provenanceTools & infrastructureRank 18 Generative and network models predict and perturb gene regulatory outputsML / AI methodsRank 19 LLMs applied to cell annotation, TF regulatory inference, and bioinformatics QAML / AI methodsRank 20 Learned models infer homology, gene presence and design constraints from genomic sequencesGenomics
Show 13 moreShow less

Sponsors