Demux and cell numbers

[clairebio27]

Hi,
Thank you very much for this method for cell hashing demux.
I am doing tcr sequencing with cell hashing . and it gives separate fastqs for HTOs and TCR (I have used cellranger and Illumina). I am writing down the method I have used to demux it using MULTI-seq, please let me know I am in the right track.

1. I did a VDJ assembly using cell ranger to get the clonotypes
2. generated count matrix for HTO data using cell ranger.
3. Used Seurat and MULTISeqDemux function to de-convolute the samples. (quantile = 0.9,, autoThresh = TRUE, qrange = seq(from = 0.1, to = 0.9, by = 0.05),)
   4.downloaded the barcodes for negative, doublets and Htos using WhichCell function in Seurat.
4. Using the merge function in R, assigned the clonotypes (matching hydrogel barcodes) for the demuxed barcodes.

Now the queries are,

1. do the parameters used for demux is apt for my case?
2. I had around 1316 cells after deconvolution. but when I tried to match with hydrogel barcodes in tcr data, only 337 are matched. The unassigned includes from HTOs .what could be the reason? could it be due to the 1 mismatch in the barcode? I assume any of my step is not doing the hammimg distance 1 correction for barcodes?

Any feedback in this regards will be appreciated.

Thanks,
Claire