Error when input data is a dataframe or tibble

[arisp99]

Bug Description

compute_coi and optimize_coi fail to run when the data is either a data frame or a tibble. They require data to be a matrix. This error is caused when our data is processed to be fed into the algorithims.

Data Frame

library(coiaf)

data_df <- data.frame(
  check.names = FALSE,
  row.names = c("FP0008-C","FP0009-C","FP0015-C","FP0016-C","FP0017-C"),
  `Pf3D7_01_v3-93378` = c(0, 0, 0, 0, 0),
  `Pf3D7_01_v3-93901` = c(0.882353, 1, 1, NA, 1),
  `Pf3D7_01_v3-94422` = c(0, 0, 0.22973, 0, 0),
  `Pf3D7_01_v3-94442` = c(0, 0, 0.30137, NA, 0),
  `Pf3D7_01_v3-94445` = c(0, 0, 0.3375, NA, 0)
)
str(data_df)
#> 'data.frame':    5 obs. of  5 variables:
#>  $ Pf3D7_01_v3-93378: num  0 0 0 0 0
#>  $ Pf3D7_01_v3-93901: num  0.882 1 1 NA 1
#>  $ Pf3D7_01_v3-94422: num  0 0 0.23 0 0
#>  $ Pf3D7_01_v3-94442: num  0 0 0.301 NA 0
#>  $ Pf3D7_01_v3-94445: num  0 0 0.338 NA 0

wsaf_df  <- data_df[1, ]
plaf_df <- colMeans(data_df, na.rm = T)
input_df <- tibble::tibble(wsaf = wsaf_df, plaf = plaf_df) %>% tidyr::drop_na()
compute_coi(input_df, "real", coi_method = "1")
#> Error: wsaf must be a non-recursive vector

Tibble

data_tb <- tibble::tibble(data_df)
str(data_tb)
#> tibble [5 × 5] (S3: tbl_df/tbl/data.frame)
#>  $ Pf3D7_01_v3-93378: num [1:5] 0 0 0 0 0
#>  $ Pf3D7_01_v3-93901: num [1:5] 0.882 1 1 NA 1
#>  $ Pf3D7_01_v3-94422: num [1:5] 0 0 0.23 0 0
#>  $ Pf3D7_01_v3-94442: num [1:5] 0 0 0.301 NA 0
#>  $ Pf3D7_01_v3-94445: num [1:5] 0 0 0.338 NA 0

wsaf_tb  <- data_tb[1, ]
plaf_tb <- colMeans(data_tb, na.rm = T)
input_tb <- tibble::tibble(wsaf = wsaf_tb, plaf = plaf_tb) %>% tidyr::drop_na()
optimize_coi(input_tb, "real", coi_method = "1")
#> Error: wsaf must be a non-recursive vector

^{Created on 2021-09-15 by the reprex package (v2.0.1)}