What am I looking at?
Each point is one bulk RNA-seq profile from ARCHS4: its learned-identity sample summary token, reduced to two dimensions. Samples that the model represents similarly sit close together.
This map contains the exported exp8 samples, balanced across labels and studies; it is not the whole training corpus. Tissues and GEO study groups are annotations added after training. Nearby samples may share biology, technical effects, or both.
Hovera point to read its metadata.
Clickto pin it and see its ten nearest neighbours in the model’s space.
Legendclick a class to isolate it, then “Show projects” to colour its GEO study groups.
Searcha GSM, GSE, tissue or disease to light up matching samples.
