-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Suggestion: Tweaked or extended comparison to DropletUtils::hashedDrops()
, perhaps in OSCA?
#5
Comments
Interesting! I have not explored the different parameters of Regarding the OSCA online book, I think that would be great! The chapter could feature both methods on the real cell line mixture dataset, which I left out in the demuxmix vignette. If the other book co-authors agree that adding demuxmix makes sense, I am happy to help if there is anything I can contribute. |
That would be great! |
Thank you for adding the comparison to
DropletUtils::hashedDrops()
in the vignette.Something I've found with applying
hashedDrops()
to a number of datasets now is that, unsurprisingly, tweaking the parameters can substantially improve the results, especially for 'low quality' HTOs.In particular, reducing the
confident.min
parameter seems to often help for these datasets.This can also be seen in a tweaked version of what you have in the vignette (I also added a comparison of each method's results to the ground truth).
Upon changing
min.confident = 2
(default) tomin.confident = 1
, the 'uncertain' droplets fromhashedDrops()
are all almost all correctly re-assigned asHTO_3
.Created on 2022-09-05 with reprex v2.0.2
Session info
I think using the default value, as you've done in the vignette, is reasonable and fair.
I guess you could add a small note to the vignette, but I can see how comparisons to other methods soon get beyond the scope of a vignette.
But it could be a great fit for the 'Orchestrating Single Cell Analysis with Bioconductor' (OSCA), in the 'Demultiplexing cell hashes' section.
It would also be a good way to promote demuxmix to new users.
I'm one of the maintainer's of the OSCA book (although not that particular section) and could probably help facilitate this if it's of interest.
The text was updated successfully, but these errors were encountered: