trans_diff with lefse method returns Negative numbers of LDA

[Irisescat]

Hi Chi,

I have an issue with the LDA value of the trans_diff with lefse method.

I used following scripts

library(microeco)
library(r2excel)
t1 <- readRDS("meco_t1.RDS")
lefse_diff <- trans_diff$new(dataset = t1, method = "lefse", group = "Group", p_adjust_method = "none")
xlsx.writeFile(lefse_diff$res_diff,sheetName = "LEfSe_result","LEfSe_result.xlsx",row.names = F)

I try to figure out, and test with trans_diff source code. It seems like this step %>% abs doesn't work well. https://github.com/ChiLiubio/microeco/blob/094994c42dbb251773f0fb67251eb013c5757e29/R/trans_diff.R#L457C9-L457C46

Thanks, Sonia

meco_t1.zip

R version 4.2.0 (2022-04-22)
Platform: x86_64-conda-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)

Matrix products: default
BLAS/LAPACK: /home/sonia/miniconda3/envs/r_4.2.0/lib/libopenblasp-r0.3.21.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=en_US.UTF-8
 [9] LC_ADDRESS=en_US.UTF-8        LC_TELEPHONE=en_US.UTF-8
[11] LC_MEASUREMENT=en_US.UTF-8    LC_IDENTIFICATION=en_US.UTF-8

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

other attached packages:
[1] r2excel_1.0.0   xlsx_0.6.5      microeco_0.19.0

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.10        pillar_1.9.0       compiler_4.2.0     RColorBrewer_1.1-3
 [5] plyr_1.8.8         tools_4.2.0        digest_0.6.31      lifecycle_1.0.3
 [9] tibble_3.2.1       nlme_3.1-162       gtable_0.3.3       lattice_0.21-8
[13] mgcv_1.8-42        pkgconfig_2.0.3    rlang_1.1.1        Matrix_1.5-4.1
[17] igraph_1.4.3       cli_3.6.1          parallel_4.2.0     rJava_1.0-6
[21] stringr_1.5.0      dplyr_1.1.2        cluster_2.1.4      xlsxjars_0.6.1
[25] generics_0.1.3     vctrs_0.6.2        grid_4.2.0         tidyselect_1.2.0
[29] glue_1.6.2         data.table_1.14.8  R6_2.5.1           fansi_1.0.4
[33] ggplot2_3.4.2      reshape2_1.4.4     magrittr_2.0.3     scales_1.2.1
[37] MASS_7.3-60        splines_4.2.0      permute_0.9-7      ape_5.7-1
[41] colorspace_2.1-0   utf8_1.2.3         stringi_1.7.6      munsell_0.5.0
[45] vegan_2.6-4

[ChiLiubio]

Hi Sonia,

Thanks very much! It is a bug! The negative values may occure when the abs fail to work. I should not use pipe operator here as it is a simple expression. I have updated the package in github. The formal version v0.20.0 will be released in CRAN in the next days. Thanks again for your finding!

Best regards, Chi

[finally-jay]

Dear Chi,

I still got negative value when using version v1.0.0, do you think I should switch back to v0.20.0 ?

[ChiLiubio]

Hi. Do you mean it is fine with v0.20.0, but negative value occures in v1.0.0?