Back

Computational array reconstruction with accumap for spatial transcriptomics

Oakes, C. G.; Pachter, L.

2026-01-09 bioinformatics
10.64898/2025.12.19.695576 bioRxiv
Show abstract

Several optics-free spatial transcriptomics assays that make use of diffusion of reads from beads have recently been developed, and provide a proof of principle for high-throughput, untargeted, spatial transcriptomics. These approaches leads to a challenging computational array reconstruction problem. We show that an statistical inference method, which we term accumap, outperforms the current approach which is to run UMAP on the bead count matrix. The accumap method makes use of predicted distances to reconstruct bead positions and can improve the accuracy of reconstruction by a factor of 3 in the diffusible Slide-Tag simulation framework.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.