breaking lapply for a blank file R

Question

I am trying to check percentage completion of common elements (BGC**) in different sample files. My input file format are as follows:

file1.txt
-----------

contig SRR5947942_idxstats.txt
BGC0000972 0
BGC0000972 0
BGC0000972 0
BGC0000972 1
BGC0000972 0
BGC0000972 0

file2.txt
----------
contig SRR5947963_idxstats.txt
BGC0000581 0
BGC0000581 22
BGC0000581 60
BGC0000581 0
BGC0000972 14
BGC0000972 24

I save them in a directory and run my script as:

filenames <- list.files(full.names=F, pattern=".txt")
output <-lapply(filenames,function(i){
  t<-read.csv(i, header=T, check.names = F, sep = " ")
  t$gene_count<-1
  t[,2][t[,2]>0]<-1
  presence_absence_df<-aggregate(. ~ contig, t, sum)
  presence_absence_df$sample_name<-names(t[2])
  colnames(presence_absence_df)<-c("BGC_Accession","Gene_presence", "Gene_count", "Sample_name")
  presence_absence_df$Percentage<-(presence_absence_df$Gene_presence/presence_absence_df$Gene_count)*100
  presence_absence_df<-presence_absence_df[presence_absence_df$Percentage != 0, ]
  presence_absence_df$tp_step2_100_percent<-length(presence_absence_df$Percentage[presence_absence_df$Percentage>=100])
  presence_absence_df<-presence_absence_df[presence_absence_df$Percentage >= 100, ]
  presence_absence_df<-data.frame(presence_absence_df)
  presence_absence_df <- subset(presence_absence_df, select = -c(Gene_presence, Gene_count, Percentage) )
  colnames(presence_absence_df)<-c("BGC_name", "Sample", "BGCs_step2_100_percent")
  presence_absence_df <- presence_absence_df [c("Sample", "BGCs_step2_100_percent", "BGC_name")]
})
Step2_results2_100<-do.call(rbind,output)

The problem is, if any of the input file has all zero, the code show error. For example, if I change the file1.txt as follows:

file1.txt
-----------

contig SRR5947942_idxstats.txt
BGC0000972 0
BGC0000972 0
BGC0000972 0
BGC0000972 0
BGC0000972 0
BGC0000972 0

Then I get:

Error in `$<-.data.frame`(`*tmp*`, "tp_step2_100_percent", value = 0L) : 
  replacement has 1 row, data has 0

I want to bypass the processing of the all zero files without showing the error. Thank you for your help!

Answer 1

Return NULL if all the values in second column is 0.

output <-lapply(filenames,function(i) {
  t <- read.csv(i, header=T, check.names = F, sep = " ")
  if(all(t[[2]] == 0)) return(NULL)
  t$gene_count<-1
  t[,2][t[,2]>0]<-1
  #Rest of the code
  #Rest of the code
})

Step2_results2_100 <- do.call(rbind,output)

When you do do.call(rbind,output) those NULL values will be ignored.