search

Zhangxf-ccnu / ENGEP

MIT License
1 stars 0 forks source link

error run engep_predict #1

Open LenonaFan opened 6 months ago

LenonaFan commented 6 months ago

Hi,

It seems that the newest version of propr (v5.0.1) doesn't have phis function anymore. So this will cause the error when I try to run engep_predict.

And also the two similarity measures "phi_s" and "rho_p" will somehow create NA value then fail to finish the analysis.

Could you please check the source code and fix this two bugs here?

Thanks a lot!

st-yang97 commented 5 months ago

Hi, Thank you for reaching out and bringing this issue to our attention. We appreciate your interest in the ENGEP package. Based on the error message and the details you provided, it seems like the problem might be related to an upgrade in the propr package (https://github.com/tpq/propr), affecting the propr::phi and propr::perb functions.

Here are some recommendations:

  1. Sub-Reference Size: Ensure that the sub-reference size is adequately chosen, avoiding sizes that are too small (e.g., 10 cells and 50 cells), as the default maximum k value in kNN regression is set to 50. We recommend utilizing the default value of n0=8000 when partitioning the reference dataset.
  2. Error Cause: The error is likely due to the recent upgrade in the propr package. We acknowledge this issue and plan to release an update for the ENGEP package to address it promptly.
  3. Potential Solutions:
    • Install a previous version of the propr package (e.g., version ‘4.2.6’) until the ENGEP package is updated.
    • If installation issues persist, you can proceed by using the remaining eight similarity measures, excluding the two in the propr package, though this might result in a slight loss of accuracy. 
      engep_predict <- engep_predict(osmfish,ref_list,pre_genes,parallel=T,get_baes=T,get_weight=T,
                       simi_list = c("pearson", "spearman", "cosine", "jaccard", "weighted_rank",
                                     "manhattan", "canberra", "euclidean"))
  4. Package Versions as a Reference:

R version 4.1.1 (2021-08-10) Platform: x86_64-conda-linux-gnu (64-bit) Running under: Ubuntu 18.04.5 LTS

Matrix products: default BLAS/LAPACK: /home/zhangxf/anaconda3/envs/yangst/lib/libopenblasp-r0.3.18.so

locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8 [7] LC_PAPER=en_US.UTF-8 LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages: [1] stats graphics grDevices utils datasets methods base

other attached packages: [1] Matrix_1.3-4 ENGEP_1.0

loaded via a namespace (and not attached): [1] compiler_4.1.1 parallel_4.1.1 grid_4.1.1 lattice_0.20-45 library(propr) packageVersion("propr") [1] ‘4.2.6’

We appreciate your patience, and we're actively working on resolving this issue. Feel free to reach out if you have further questions or concerns.