Genomic loci that match peptides can be identified using the following steps:

1. Load in the libraries and the data:

library(MSnID)
library(dplyr)
library(Biostrings)

msnid <- MSnID() # create object
msnid <- read_mzIDs(msnid, file.path(getwd(), "datasets", "ch6", "PXD006247.mzXML.mzid")) 

peptide_info <- as(msnid, "data.table") %>%
 filter(isDecoy == FALSE) %>%
 select(spectrumID, pepSeq, ) %>%
 mutate(fasta_id = paste0( spectrumID, ":", 1:length(spectrumID)) )

2. Extract the peptide sequence and save it as a fasta file:

string_set <- AAStringSet(peptide_info$pepSeq )
names(string_set) <- peptide_info$fasta_id
writeXStringSet(string_set[1], file.path(getwd(), "datasets", "ch6", "peptides.fa"))

3. Prepare the filenames for the BLAST run:

input_seqs <- file.path(getwd(), "datasets", "ch6", "peptides.fa")
genome_seqs <- file.path(getwd(), "datasets", "ch6", "ecoli_genome.fasta")
output_blast <- file.path(getwd(), "datasets", "ch6", "out.blast")

4. Prepare the BLAST command:

command <- paste0(
 "tblastn", 
 " -query ", input_seqs ,
 " -subject ", genome_seqs, 
 " -out ", output_blast,
 " -word_size 2 -evalue 20000 -seg no -matrix PAM30 -comp_based_stats F -outfmt 6 -max_hsps 1"
 )

5. Run BLAST as a background process:

library(withr)
with_path("/Users/macleand/miniconda2/bin", system(command, wait = TRUE) )

6. Convert BLAST into GFF and GRanges:

results <- read.table(output_blast)

blast_to_gff <- function(blst_res){
 blst_res %>% 
 mutate(
 seqid = V2,
 source = rep("tblastn", length(V1)),
 type = rep(".", length(V1)),
 start = V9,
 end = V10,
 score = V3,
 strand = rep(".", length(V1)),
 phase = rep(".", length(V1)),
 attributes = paste("Name=",V1)
 ) %>%
 select( - starts_with("V") )
}

gff_df <- blast_to_gff(results)

library(GenomicRanges)
granges<-makeGRangesFromDataFrame(gff_df)