SCEasy outputs different OBS from emptydroips output


I am currently analysing my single cell RNAseq data. I have 4 plates of comparable data I want to analyze together following the tutorial “Generating a single cell matrix using Alevin”.

It works, if I analyze the plates individually but when I try to concatenate the data errors occur. This is because the SCeasy convert gives different outputs. For 2 of my plates I get 7 obs (Barcode, emptyFDR, emptyLimited, emptyLogProb, emptyPValue, emptyTotal, index), the other 2 only 5 obs (Barcode, emptyFDR, emptyLogProb, emptyTotal, index).

So when I concatenate the data into 1 big AnnData object, tools like Scanpyfiltercells dont work, cause some data are missing the 2 obs. Does anyone have an idea, why SCEasy “loses” the obs in 2 cases? The tabular output of emptydrops has al 7 obs for all plates so the input is similar.

You can find the History here:


The SCEasy steps with different outputs for the 4 plates are 330-334 with the inputs (from emptydrops) at 322-329.

Thanks for your help!