Make a bar chart based on genes recurrence across several data frames

Question

I applied an analysis across 17 cancer datasets. This analysis extracts the most important genes in those datasets and stores them in dataframes, along with p-values and log fold change. Since I have 17 datasets, I have 17 different dataframes containing those genes. The gene names are stored as rownames. I'll dput an example.

I want to make a bar chart, so that the names of the reccurent genes would be in the x axis, and the y axis shows in how many datasets those genes appear.

I drew an example of what I want more or less, it's easier to understand what I'm trying to do. Here, GeneA appears in two datasets (out of the 17), GeneB appears in five, and so on..

And this is a sample, from one of the dataframes containing the important genes:

structure(list(logFC = c(0.685619610367974, 1.02911399977095, 
1.21579832342335, 0.841748375326278, -0.95542189916311, -0.622394818631047, 
-1.09930985256326, -0.690554417934163, -1.07611805007749, -0.948645964224887, 
-1.19079386783244, -0.756831495690092, -1.31779396563897, -0.924931124917402, 
0.643565666416337, -1.53363592483325, -1.32984685582547, -1.51833806458711, 
-1.27482926851119, -0.646054648595126, 0.730157851421398, -1.35025872065137, 
-0.541002828858919, 0.99837288668516, -1.5943393059336, -2.98196115973334, 
-1.12356324822189, 0.824177500735602, -0.474657933885741, -1.03495843655562
), AveExpr = c(5.05747198914854, 7.75940716840556, 3.8862859573242, 
6.45085792154631, 4.43329710921497, 4.69052583450353, 5.22118708978412, 
4.01613055295418, 4.69525023673379, 5.09372142578403, 5.14270112479785, 
3.63232072961097, 5.69808136458236, 4.15064883311976, 6.26413864971108, 
2.5581138769371, 2.90910253040127, 2.41800584482312, 3.43553185874516, 
7.03837991758644, 6.46936113235691, 2.67724307365115, 4.99684831009086, 
6.82992552495244, 1.30882233740874, -2.28529840100659, 3.07710437824841, 
6.28691890035385, 5.2405192106965, 2.5561726432035), t = c(4.20511802882062, 
3.97430617985228, 4.18461388788438, 3.75065833089504, -3.74904828469963, 
-3.69125356189902, -3.63103195286571, -3.59655040997674, -3.52954054833831, 
-3.4712282871982, -3.45812015299906, -3.52003442282881, -3.44172287211435, 
-3.46589605541631, 3.42058503629755, -3.54504837781242, -3.52836396936311, 
-3.54667074361479, -3.41439156856387, -3.34450445379578, 3.33286553107132, 
-3.46154088027199, -3.32150382065004, 3.30509454726963, -3.65100886393021, 
-4.47839052350898, -3.3738507671459, 3.25825505673401, -3.2593854571864, 
-3.41466969858952), P.Value = c(0.000106021402821303, 0.000223806789211493, 
0.000113375581956081, 0.000453594412082253, 0.000455876454295666, 
0.000545499225012084, 0.000656761966844333, 0.000729929667865844, 
0.000894998135798265, 0.00106709947324608, 0.00110991114624083, 
0.000921121368589992, 0.00116576947504538, 0.00108432150993663, 
0.00124172990454849, 0.000853893040602941, 0.000898192669499512, 
0.000849698463021537, 0.0012648632706601, 0.00155592601730215, 
0.00161018547122856, 0.00109858387680183, 0.00166487844680117, 
0.00174698324894766, 0.000617640897128276, 4.28624781124686e-05, 
0.00142670855222148, 0.00200294162363801, 0.00199636736769239, 
0.00126381569483115), adj.P.Val = c(0.491114677138253, 0.553985748095362, 
0.491114677138253, 0.642128989508303, 0.642128989508303, 0.642128989508303, 
0.642128989508303, 0.642128989508303, 0.642128989508303, 0.642128989508303, 
0.642128989508303, 0.642128989508303, 0.642128989508303, 0.642128989508303, 
0.642128989508303, 0.642128989508303, 0.642128989508303, 0.642128989508303, 
0.642128989508303, 0.642128989508303, 0.642128989508303, 0.642128989508303, 
0.642128989508303, 0.642128989508303, 0.642128989508303, 0.491114677138253, 
0.642128989508303, 0.642128989508303, 0.642128989508303, 0.642128989508303
), B = c(0.864951243823227, 0.43558162271344, 0.179475533728334, 
-0.162935279147593, -0.230345274467081, -0.364440984772641, -0.468609367293149, 
-0.680242692703549, -0.74700189253372, -0.868687025452187, -0.897619983658367, 
-0.923461454055463, -0.930910276805026, -0.959121075652368, -0.990427578949439, 
-0.995826555677875, -1.00570839987277, -1.09301618043057, -1.14138893444917, 
-1.166925812266, -1.19887366135859, -1.20643698734222, -1.24279579981017, 
-1.26418840296138, -1.26951050271089, -1.32756343917141, -1.34169073363162, 
-1.38132958268032, -1.38305582586065, -1.40684725754031)), row.names = c("FLCN", 
"PMP22", "MEIS3P1", "ABCC5", "ZBTB18", "MTM1", "DOCK11", "TNFAIP8L1", 
"MMD", "HIVEP2", "ANTXR2", "PHLPP2", "BMS1P20", "NMT2", "TCEAL9", 
"IL2RA", "MAP4K1", "NR2F1", "CD28", "UBE2J1", "ENTPD6", "AMIGO2", 
"RNGTT", "KIF13A", "BMP6", "GPR88", "HDAC9", "MARVELD1", "STYX", 
"DGKE"), class = "data.frame")

NOTE - there might be too many reccurent genes. In that case, is there any way to pick the top 20 genes for example? like the 20 genes that appear the most across those dataframes.

Thanks!!

user_3732271 · Answer

Assuming there are multiple datasets created in the global environment, place them in a list, loop over the list with map, convert the row.names to column, bind to a single dataset, and subset the top 20 with slice_max after creating a frequency column with add_count

library(dplyr)
library(tibble)
library(purrr)
list(df1, df2, ..., dfn) %>% 
  map_dfr(~ .x %>% 
       rownames_to_column("Genes")) %>% 
  add_count(Genes) %>% 
  slice_max(n = 20, order_by = Genes)

Make a bar chart based on genes recurrence across several data frames

And this is a sample, from one of the dataframes containing the important genes:

NOTE - there might be too many reccurent genes. In that case, is there any way to pick the top 20 genes for example? like the 20 genes that appear the most across those dataframes.

1 Answers