bm2-lab / sclearn Goto Github PK
View Code? Open in Web Editor NEWscLearn:Learning for single cell assignment
scLearn:Learning for single cell assignment
Dear authors,
When I ran the codes that you shared in the github, what I got are as follows:
"[1] "The number of missing features in the query data is 79 "
[1] "The rate of missing features in the query data is 0.0671768707482993 "'
and the reference dataset is deng-reads.rds, the query dataset is zeisel.rds.
But I can't get the accuracy that you provided in the papers.
Like this "deng_zeisel scLearn 0.996"
Is something missing when I ran the codes.
I will appreciate it if you can helo me out of this. Thank you for your time.
I ran the codes that you provided in github, and I got the scLearn_predict_result. But I have a problem about the the meaning of additional information in predict_result. I have read the codes that how to get additional information, still have no clues. What'more, I can't find the types of query cells, so I don't know how to get the accuracy of scLearn. You don't provide the types of query cells and the method of how to get the accuracy. I really hope to receive your reply. I will appreciate it if you can solve my problems.
It is not possible to download the pre-trained models! It requires to create an account, verify it by the passcode sent by SMS, but each time the process fails with no explanation whatsoever.
Dear authors,
I find it is hard to distinguish some cell types if they are similar in the reference dataset using default parameters. There are too many 'unassigned' in the output. I wonder how to adjust the parameters in scLearn_model_learning() to get a less strict model? Thanks.
Hi, When I used the command devtools::install_github("bm2-lab/scLearn"), I got the bug information:
Installing package into ‘C:/Software/R/Rlibrary’
(as ‘lib’ is unspecified)
* installing *source* package 'scLearn' ...
** using staged installation
** R
** byte-compile and prepare package for lazy loading
Error: (converted from warning) package 'dml' was built under R version 4.0.3
Execution halted
ERROR: lazy loading failed for package 'scLearn'
* removing 'C:/Software/R/Rlibrary/scLearn'
I have checked my R version which was version 4.0.2 and the R version for package more than 3.6.1. By the way I also reinstalled the dml package, but it came to failed again. Have you ever met this bug or how could I fix it? THX
Could you kindly release the preprocessed scRNA-seq dataset.
I download the PBMC dataset from "https://singlecell.broadinstitute.org/single_cell/study/SCP424/single-cell-comparison-pbmc-data#study-download", but I found some cell type labels (InDrop2 and pbmc1_10x_v2 ) are missing in meta.txt.
Hello, I encounter a error when run the:
scLearn_predict_result<-scLearn_cell_assignment(scLearn_model_learning_result,data_qc_query$expression_profile).
The message is:
[1] "The number of missing features in the query data is 0 "
[1] "The rate of missing features in the query data is 0 "
Error in scLearn_model_learning_result$trans_matrix_learned[[r]] %*% expression_profile_query_hvg :
non-conformable arguments
could you give me some advice for this error? The R version is 4.0.2.
Dear authors,
When I ran this code"scLearn_predict_result<-scLearn_cell_assignment(scLearn_model_learning_result,data_qc_query$expression_profile,diff=0.05,threshold_use=TRUE,vote_rate=0.6)", there was a error:
Error in scLearn_cell_assignment(scLearn_model_learning_result, data_qc_query$expression_profile, :
参数没有用(threshold_use = TRUE).
Even if I changed the value of threshold_use to FALSE, it still has problem.
Can you explain to me that why this happenen? Thank you for your time.
Hello, I encounter a error when run the Cell assignment part:
data2<-readRDS('pancreas_human_segerstolpe.rds')
rawcounts2<-assays(data2)[[1]]
Error in (function (classes, fdef, mtable) :
unable to find an inherited method for function ‘assays’ for signature ‘"list"’
The pancreas_human_segerstolpe.rds file was downloaded from pre-trained scLearn models of the 30 datasets.
How should I deal with such error? Looking forward to your reply .
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.