CITATION.cff

cff-version: 1.2.0
message: "If you use this software in academic work, please cite it as below."
title: SpatialTranscriptFormer
abstract: >-
  A transformer framework that predicts spatially-resolved biological pathway
  activity directly from H&E histology. Learnable pathway tokens cross-attend to
  histology patch features (a configurable quad-flow interaction), and a
  dot-product + Softplus head produces per-spot pathway-activity scores supervised
  against offline-computed targets.
type: software
authors:
  - family-names: Wilson
    given-names: "Benjamin Isaac"
version: "0.1.0"
license: Apache-2.0
repository-code: "https://github.com/BenjaminIsaac0111/SpatialTranscriptFormer"
url: "https://github.com/BenjaminIsaac0111/SpatialTranscriptFormer"
keywords:
  - spatial transcriptomics
  - computational pathology
  - deep learning
  - pathway activity
  - histology
  - transformer