Author

Xiang Lin

Harvard UniversityORCID 0000-0003-2789-4449Unclaimed ORCID profile · follow to get notified of new preprints

Additional preprints under this name without ORCID

Xiang Lin1 paper

Biology Disciplines

Click to filter author papers

Bioinformatics· 2

Signature Assays, Organisms & Methods

Click to filter · or open tag

Approach:deep learning· 2 Research Focus:perturbation response prediction· 2 Research Focus:benchmarking· 1 Research Focus:drug response / drug discovery· 1 Approach:graph neural network· 1 Organism:mouse· 1 Assay:spatial transcriptomics· 1 Approach:variational autoencoder· 1

Frequent Co-Authors

Marinka ŽitnikiD· 2 shared Boyang FuiD· 1 shared Guadalupe GonzaleziD· 1 shared Isuru S. HerathiD· 1 shared Kirill VeselkoviD· 1 shared M KellisiD· 1 shared Michael M. BronsteiniD· 1 shared Soumya Ghosh· 1 shared

Activity by Year

2025-2025 · peak 2/yr

Co-Author Collaboration Network

9 top collaborators · 20 mutual co-author links

Click any collaborator node to inspect or filter shared papers

Focal author (center)Selected co-authorORCID co-authorUnlinked name

Curved chords = mutual co-authorship

2 shared papersBioinformatics

Marinka ŽitnikORCID

Mutual co-authors in cluster

Author profile →

2 papers

Recent·Most cited

BioinformaticsbioRxivPreprintNov 8, 2025

Generative modeling of spatially resolved perturbation responses with CONCERT

Xiang Lin, Zhenglun Kong, Boyang Fu…·Marinka Žitnik

CONCERT, a niche-aware generative model, predicts spatial perturbation responses more accurately than existing methods across multiple tissues and experimental setups.

4 citations

mousespatial transcriptomics

𝕏 PostBluesky

BioinformaticsNature Biomedical EngineeringPeer-reviewedSep 9, 2025

Combinatorial prediction of therapeutic perturbations using causally-inspired neural networks

Guadalupe Gonzalez, Xiang Lin, Isuru S. Herath…·Marinka Žitnik

PDGrapher predicts combinatorial therapeutic targets to reverse disease phenotypes using a causally inspired GNN, outperforming existing methods in cell line and genetic perturbation datasets.

35 citationstop 1%

graph neural networkdeep learning

𝕏 PostBluesky