Hi,

I am working with the RNA-Seq dataset (one sample per condition per subject). There are a total of 6 subjects, of which 3 are unrelated healthy subjects (each healthy subject with 3 treatments + 1 untreated) and 1 patient (3 treatments + 1 untreated) + 2 relatives (3 treatments + 1 untreated). I calculated log2FC values for each subject and treatment condition relative to the untreated condition of the same subject using edgeR. I subsetted and combined the log2FC of each subject (now contains; Treatment 1, Treatment 2, and Treatment 3 w.r.t untreated) and have combined the data.frame (see below, example) in R.

How do I calculate and consider the genes with a |log2FC| >=1 in at least 2 out of 3 unrelated healthy subjects in vitro treatment-responsive genes. Additionally, I would like to residual response.

Note: The residual responses of the subjects were calculated based on the number of responsive genes passing the above filter in 2 out of 3 healthy subjects (= number of responsive genes in a subject / total number of responsive genes in healthy controls) × 100). Residual responses are described in (PubMed: 31784499, 34427831, 34214472, 30143481). We would normally start establishing residual response based on the three healthy subjects and then check in the patient and 2 relatives how they deviate from the normal response.

Data

    print(tail(log2FC))
        HC1_Treatment_1.logFC HC1_Treatment_2.logFC HC1_Treatment_3.logFC
Gene_15            0.07164503            0.59904274             0.0278425
Gene_16           -0.28451267            0.62414548            -0.1684542
Gene_17           -0.03066383           -0.30650261            -0.2355185
Gene_18           -0.48784162           -0.15523340            -0.1797719
Gene_19            2.00000000            0.19152352            -0.9262081
Gene_20            0.95276316            0.07279556             3.0000000
        HC2_Treatment_1.logFC HC2_Treatment_2.logFC HC2_Treatment_3.logFC
Gene_15             0.2327490            0.23485943             0.4010577
Gene_16             0.5390979            0.25679592             0.2962973
Gene_17            -0.2327766           -0.07059749            -0.3062303
Gene_18             0.1934787           -0.13324728             0.0931452
Gene_19             0.1658384            1.90000000            -0.1275337
Gene_20             4.0000000            5.60000000             0.2044699
        HC3_Treatment_1.logFC HC3_Treatment_2.logFC HC3_Treatment_3.logFC
Gene_15           -0.03304623           0.254568488            0.27022068
Gene_16           -0.21044532           0.407651132            0.07366016
Gene_17            0.10053437          -0.000781975            0.18696837
Gene_18           -0.11581915           0.069750175            0.20076677
Gene_19           -0.82945255           0.218905487            5.00000000
Gene_20            0.16254931           4.000000000           -0.35071090
        P1_Treatment_1.logFC P1_Treatment_2.logFC P1_Treatment_3.logFC
Gene_15           -0.2997141           -0.5171820         -0.466957656
Gene_16            0.3445041           -0.3491658         -0.010136887
Gene_17            0.9525266            0.8360276          0.255081070
Gene_18            1.0563388            0.8489729          0.530710073
Gene_19            6.0000000            0.4492499          0.009571011
Gene_20            6.0000000            0.1065795          0.624807496
        R1_Treatment_1.logFC R1_Treatment_2.logFC R1_Treatment_3.logFC
Gene_15          -0.03307696          -0.07008209          0.006218129
Gene_16           0.66286592           1.11764324          0.451419358
Gene_17          -0.09582274          -0.36306320         -0.379897186
Gene_18           0.47976491           0.42302656         -0.164057640
Gene_19           4.40000000          -0.06769135          6.800000000
Gene_20           5.00000000          -0.50945501          0.060202575
        R2_Treatment_1.logFC R2_Treatment_2.logFC R2_Treatment_3.logFC
Gene_15            0.3330158           0.05105085           0.07517517
Gene_16            0.5680208           0.39302658           0.20795389
Gene_17            0.1331289          -0.65545389          -0.27831143
Gene_18            0.2029048          -0.19068632           0.27433957
Gene_19            0.5384385          -0.98295489          -0.05461810
Gene_20            6.0000000          -0.29869977          -0.38437364

Code: filter criteria |log2FC| >= 1

# User-defined function to do the custom filtering:
my_filter <- function(xx) {
  xx <- xx[!is.na(xx)] 
  sum(ifelse((xx >= 1) | (xx <= -1), TRUE, FALSE))
}

# Do it across all `Healthy` subject columns with "treatments" in the column name
library(dplyr)
log2FC %>% 
  summarise(across(contains("HC"), my_filter)) %>% 
  as.data.frame(row.names="counts") -> log2FC_v1

Best Regards,

Toufiq