Geneformer is a foundation transformer model pretrained on a large-scale corpus of ~30 million single cell transcriptomes to enable context-aware predictions in settings with limited data in network biology. Here, we will demonstrate a basic workflow to work with ***Geneformer*** models. These notebooks include the instruction to: 1. Prepare input datasets 2. Finetune Geneformer model to perform specific task 3. Using finetuning models for cell classification and gene classification application

Single-cell RNA data allows cell-cell communications (***CCC***) methods to infer CCC at either the individual cell or cell cluster/cell type level, but physical distances between cells are not preserved Almet, Axel A., et al., (2021). On the other hand, spatial data provides spatial distances between cells, but single-cell or gene resolution is potentially lost. Therefore, integrating two types of data in a proper manner can complement their strengths and limitations, from that improve CCC analysis. In this pipeline, we analyze CCC on Visium data with single-cell data as a reference. The pipeline includes 4 sub-notebooks as following 01-deconvolution: This step involves deconvolution and cell type annotation for Visium data, with cell type information obtained from a relevant single-cell dataset. The deconvolution method is SpatialDWLS which is integrated in Giotto package. 02-giotto: performs spatial based CCC and expression based CCC on Visium data using Giotto method. 03-nichenet: performs spatial based CCC and expression based CCC on Visium data using NicheNet method. 04-visualization: visualizes CCC results obtained from Giotto and NicheNet.

Recent technological advancements have enabled spatially resolved transcriptomic profiling but at multi-cellular pixel resolution, thereby hindering the identification of cell-type-specific spatial patterns and gene expression variation. To address this challenge, we develop STdeconvolve as a reference-free approach to deconvolve underlying cell types comprising such multi-cellular pixel resolution spatial transcriptomics (ST) datasets. Using simulated as well as real ST datasets from diverse spatial transcriptomics technologies comprising a variety of spatial resolutions such as Spatial Transcriptomics, 10X Visium, DBiT-seq, and Slide-seq, we show that STdeconvolve can effectively recover cell-type transcriptional profiles and their proportional representation within pixels without reliance on external single-cell transcriptomics references. **STdeconvolve** provides comparable performance to existing reference-based methods when suitable single-cell references are available, as well as potentially superior performance when suitable single-cell references are not available. STdeconvolve is available as an open-source R software package with the source code available at https://github.com/JEFworks-Lab/STdeconvolve .