Analyze_Sanger_Functions.R

##----------------------------Sanger_Analysis_Functions for Sanger Sequencing reads----------------------------------------

if (!requireNamespace(c("librarian", "BiocManager")))
  install.packages(c("librarian", "BiocManager"))
library("librarian")
library("BiocManager")

librarian::shelf(devtools, tidyverse, lib = tempdir(), update_all = TRUE)

BiocManager::install("sangeranalyseR", update = FALSE)
library(sangeranalyseR)

#--------------------------------------------load files function------------------------------------------------------------
#load abi files-------------------------------------------------------------------------------------------------------------
#rename files to remove date and time stamp---------------------------------------------------------------------------------
#index forward and reverse reads and create groups--------------------------------------------------------------------------

load.files <- function(path = path){
  
  #get the file names and the list of file names you want to replace them with
  #assumes file names are in the following convention: YYYY_MM_DD_16S_ACC#_FWD_WELL_DATE_STAMP.ab1
  old_file_names <- dir(path, pattern=".ab1", full.names = TRUE)
  new_file_names<- gsub("FWD_\\w+\\d+_\\d+_\\d+.*", "FWD.ab1", old_file_names)
  new_file_names<- gsub("REV_\\w+\\d+_\\d+_\\d+.*", "REV.ab1", new_file_names)
  
  file.rename(from = old_file_names, to = new_file_names)
  
  #Index forward and reverse reads
  files_cleaned = new_file_names
  f_matches = str_match(files_cleaned, "FWD")
  f_indices = which(!is.na(f_matches))
  r_matches = str_match(files_cleaned, "REV")
  r_indices = which(!is.na(r_matches))
  
  #ONLY keep files that match either FWD or REV suffix
  keep = c(f_indices, r_indices)
  files_cleaned = files_cleaned[keep]
  
  # remove the suffixes to create a groupname
  files_cleaned = gsub("_FWD.*", "", files_cleaned)
  files_cleaned = gsub("_REV.*", "", files_cleaned)
  
  #create groups
  group_dataframe =  data.frame("file.path" = new_file_names[keep], "group" = files_cleaned)
  groups = unique(group_dataframe$group)
  
  return(groups)
}


#----------------------------------------CB.Contig function using sangercontig-----------------------------------------------
#CB.Contig function generates a contig from a fwd and reverse read using sangercontig function-------------------------------
#fasta consensus is output into fasta file-----------------------------------------------------------------------------------
#quality data for fwd and rev read are subsetted and returned as a summary object--------------------------------------------

CB.Contig<- function(path, contigName, suffixForwardRegExp, suffixReverseRegExp, file_name_fwd, file_name_rev){
  
  print("reading forward and reverse reads and generating contig")
  
  sangerContig <- SangerContig(
    inputSource           = "ABIF",
    ABIF_Directory        = path,
    contigName            = contigName,
    REGEX_SuffixForward   = suffixForwardRegExp,
    REGEX_SuffixReverse   = suffixReverseRegExp,
    TrimmingMethod        = "M2",
    M1TrimmingCutoff      = NULL,
    M2CutoffQualityScore  = 40,
    M2SlidingWindowSize   = 10,
    minReadLength         = 0,
    signalRatioCutoff     = 0.33,
    showTrimmed           = TRUE,
    geneticCode           = GENETIC_CODE)
 
  print("exporting fasta sequence")
  
  writeFasta(sangerContig, outputDir = "../Fasta_Sequences/", selection = "contig")
  
  print("exporting contig alignment")
  
  alignment = sangerContig@alignment
  alignment<- DNAMultipleAlignment(alignment)
  filepath<- file.path("../Results", paste("contigAlign", contigName, ".txt", sep=""))
  write.phylip(alignment, filepath = filepath)
  
  print("subsetting quality data")
  
  QualityFWD <- sangerContig@forwardReadList[[file_name_fwd]]@QualityReport
  QualityREV <- sangerContig@reverseReadList[[file_name_rev]]@QualityReport
  
  print("Generating read summary")
  
  read.summary = c("consensus.length"            = sangerContig@contigSeq@length,
                   "trimmed.seq.length.FWD"       = QualityFWD@trimmedSeqLength,
                   "trimmed.seq.length.REV"       = QualityREV@trimmedSeqLength,
                   "trimmed.Mean.qual.FWD"       = QualityFWD@trimmedMeanQualityScore,
                   "trimmed.Mean.qual.REV"       = QualityREV@trimmedMeanQualityScore)
  
  return(list("summary" = read.summary, "contig" = sangerContig))
  
}

#----------------------------------Summarize.Sanger function----------------------------------------------------------------
#Summarize.Sanger function runs sanger contig function and puts subsetted quality data into a dataframe---------------------

Summarize.Sanger<- function(group, path = path, summarylist = summarylist){
  
  file_name_fwd = paste(group, "_FWD.ab1", sep="")
  file_name_rev = paste(group, "_REV.ab1", sep="")
  contigName = basename(group)
  
  col_names = c("Consensus length",
                "trim length FWD",
                "trim length FWD",
                "trim MeanQual FWD",
                "trim MeanQal REV")
  
  
  consensus_sequence <- CB.Contig(path =                path,
                                  contigName =          contigName,
                                  suffixForwardRegExp = "_FWD",
                                  suffixReverseRegExp = "_REV",
                                  file_name_fwd =       file_name_fwd,
                                  file_name_rev =       file_name_rev)
  
  #separate summary data from the consensus sequence object 
  #turn it into a dataframe with row names from above
  summary = consensus_sequence$summary
  summary = t(summary) %>% 
    as.data.frame(., colnames = col_names)
  
  #add a column called sample which is equal to the contig name
  summary$sample <- contigName
  
  summarylist[[group]]<-summary
}

#---------------------------------------analyze.sequences function------------------------------------------------------
#Function to run summarize.sanger on groups of files and output  summary of quality results for all samples-------------

analyze.sequences<- function(path){
  
  #load the files and group into groups based on accession #
  groups<- load.files(path) 
  
  #generate an empty summary list to put the summary data in
  summarylist = list()
  
  #run summarize.sanger function on all files in the path to generate fasta files
  summarylist<- lapply(groups, FUN = Summarize.Sanger, path = path, summarylist = summarylist)
  #then concatenate the summary data into a single df
  summary_data<- do.call(rbind, summarylist)
  
  #change the order of the columns so the contig name is the first column of the df
  summary_data<- summary_data[, c(6,1:5)]
  
  #export the summary data into a csv in the Results folder
  Resultpath<- file.path("../Results", paste("Quality_Report", basename(path), ".csv", sep=""))
  write.csv(summary_data, file = Resultpath, row.names = FALSE)
  
}

#######################################################################################################################

#------------------------------------single.read function---------------------------------------------------------------

single.read<- function(readFileName, readFeature){
  sangerRead<- SangerRead(inputSource = "ABIF",
                          readFeature  = readFeature,
                          readFileName = readFileName,
                          geneticCode = GENETIC_CODE,
                          TrimmingMethod = "M2",
                          M1TrimmingCutoff = NULL,
                          M2CutoffQualityScore = 15,
                          M2SlidingWindowSize = 10, 
                          baseNumPerRow = 100,
                          heightPerRow = 200,
                          signalRatioCutoff = 0.33,
                          showTrimmed = TRUE)
  sangerRead
  writeFasta(sangerRead, outputDir = "../Fasta_Sequences", compress = FALSE, compression_level = NA)
  
  Quality <- sangerRead@QualityReport
  
  print("Generating read summary")
  
  read.summary = c("trimmed.seq.length"      = Quality@trimmedSeqLength,
                   "trimmed.Mean.qual"       = Quality@trimmedMeanQualityScore)
  
  return(list("summary" = read.summary, "Read" = sangerRead))
  
}

###
#----------------------------------Summarize.Single function----------------------------------------------------------------
#Summarize.Sanger function runs sanger contig function and puts subsetted quality data into a dataframe---------------------

Summarize.Single<- function(readFileName, readFeature, summarylist = summarylist){
  
  singleName = basename(readFileName)
  
  col_names = c("trim length",
                "trim MeanQual")
  
  
  single_sequence <- single.read( readFileName = readFileName,
                                  readFeature = readFeature)
  
  #separate summary data from the readsequence object 
  #turn it into a dataframe with row names from above
  summary = single_sequence$summary
  summary = t(summary) %>% 
    as.data.frame(., colnames = col_names)
  
  #add a column called sample which is equal to the contig name
  summary$sample <- singleName
  
  summarylist[[readFileName]]<-summary
}
  


#---------------------------------------analyze.single.sequence function------------------------------------------------------
#Function to run summarize.sanger on single reads and output  summary of quality results for all samples-------------

analyze.single.sequence<- function(readFileName, readFeature){
  #generate an empty summary list to put the summary data in
  summarylist = list()
  
  #run summarize.sanger function on all files in the path to generate fasta files
  summarylist<- Summarize.Single(readFileName = readFileName, readFeature = readFeature, summarylist = summarylist)
 
  
  #change the order of the columns so the contig name is the first column of the df
  summary_data<- summarylist[, c(3,1,2)]
  
  #export the summary data into a csv in the Results folder
  Resultpath<- file.path("../Results", paste("Quality_Report", basename(readFileName), ".csv", sep=""))
  write.csv(summary_data, file = Resultpath, row.names = FALSE)
  
}