txWang

Hi, Sorry about that. We have included the "write_dataset" function in "R/func_data.R". We have also changed the filename from "R/func_read_data.R" to "R/func_data.R" for less confusion.

Hi, The intuition of BERMUDA, which is using domain adaptation to align similar cell clusters across different batches, should work on CyTOF data. However, the design of the whole BERMUDA...

Hi, Thank you for your question. We apply MMD loss on pairs of similar clusters from different batches. During the construction of a mini-batch for training the network, we sample...

Thanks for the comment. Yes, it should be d2-4. Sorry for the confusion. In the case of Experiment removal1 and Experiment removal2, the performance was evaluated with 3 scores. Since...

While the use of domain adaptation in BERMUDA does not specifically restrict to single-cell data, the whole workflow of BERMUDA also depends on Seurat and Metaneighbor, which are designed for...

Hi, You could modify "main_pancreas.py" or "main_pbmc.py" to suit your own dataset given that your data has been pre-processed by Seurat for normalization and cluster identification and MetaNeighbor for cluster...

Hi, Sure. BERMUDA supports multiple datasets for batch correction. You just need to modify the code accordingly and feed the pre-processed datasets as well as the MetaNeighbor similarity matrix into...

Hi, In short, "loss_transfer" utilized mmd loss between pairs of similar clusters between different batches.

Hi, We used two packages in R and saved the results as .csv file in order to run BERMUDA. You could follow the preprocessing steps in BERMUDA/R/pre_processing.R First, we used...

Hi, Thank you for your question. Similar to many batch correction methods, BERMUDA removes batch effects by projecting the original data to a low dimensional space (dimensionality equals to 20...