The original datasets of bivalent domains
Dear authors,
Thank you for your excellent work. I currently conduct some research based on the experiments of Fig.3c, which requires the original gene expression data of PanglaoDB. I search the samples on https://panglaodb.se/samples.html by the id SRA553822 you provide in the name of the dataset file, and I find that it only contains 6501 samples. Could you provide me with the full ids of PanglaoDB for the samples you used in Fig.3c? Thanks a lot in advance.
Thank you for your interest in Geneformer. Panglao displays 6501 cells in their clustering viewer. However, if you download the data, the dataset has 35248 cells. Of these, 34254 passed our filtering criteria. We randomly selected 15000 of these for the fine-tuning task as described in the manuscript.
Thank you for your kind reply. Could you release the code for filtering the dataset? Thanks again!
Thank you for your question! Please refer to the manuscript methods for the filtering criteria.