Mercurial > repos > bornea > saint_preprocessing
comparison pre_process_protein_name_set.R @ 32:63008bdf576e draft
Uploaded
author | bornea |
---|---|
date | Tue, 26 Apr 2016 17:16:03 -0400 |
parents | e6e456d3ac14 |
children |
comparison
equal
deleted
inserted
replaced
31:761e1ad2b130 | 32:63008bdf576e |
---|---|
78 peptides_txt_mapped_log2[peptides_txt_mapped_log2 == -Inf] <- NA | 78 peptides_txt_mapped_log2[peptides_txt_mapped_log2 == -Inf] <- NA |
79 #uniprot accessions WITHOUT isoforms; it looks like only contaminants contain isoforms anyways. | 79 #uniprot accessions WITHOUT isoforms; it looks like only contaminants contain isoforms anyways. |
80 mapped_protein_uniprotonly = str_extract(peptides_txt_mapped_log2$Uniprot,"[OPQ][0-9][A-Z0-9]{3}[0-9]|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}") | 80 mapped_protein_uniprotonly = str_extract(peptides_txt_mapped_log2$Uniprot,"[OPQ][0-9][A-Z0-9]{3}[0-9]|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}") |
81 mapped_protein_uniprot_accession = str_extract(peptides_txt_mapped_log2$Uniprot,"[OPQ][0-9][A-Z0-9]{3}[0-9](-[0-9]+)?|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}(-[0-9]+)?|[OPQ][0-9][A-Z0-9]{3}[0-9]|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}") | 81 mapped_protein_uniprot_accession = str_extract(peptides_txt_mapped_log2$Uniprot,"[OPQ][0-9][A-Z0-9]{3}[0-9](-[0-9]+)?|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}(-[0-9]+)?|[OPQ][0-9][A-Z0-9]{3}[0-9]|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}") |
82 peptides_txt_mapped_log2$mapped_protein = mapped_protein_uniprotonly | 82 peptides_txt_mapped_log2$mapped_protein = mapped_protein_uniprotonly |
83 names_db = str_extract(swissprot_fasta,"[OPQ][0-9][A-Z0-9]{3}[0-9]|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}") | |
84 names_db = names_db[!is.na(names_db)] | |
83 # Runs the Tukey function returning completed table. | 85 # Runs the Tukey function returning completed table. |
84 peptides_txt_mapped_log2 = subset(peptides_txt_mapped_log2,mapped_protein %in% swissprot_fasta) | 86 peptides_txt_mapped_log2 = subset(peptides_txt_mapped_log2,mapped_protein %in% names_db) |
85 if (nrow(peptides_txt_mapped_log2) == 0) { | 87 if (nrow(peptides_txt_mapped_log2) == 0) { |
86 print("Uniprot Database does not have any of the proteins in the peptides file") | 88 print("Uniprot Database does not have any of the proteins in the peptides file") |
87 quit() | 89 quit() |
88 } | 90 } |
89 protein_intensities_tukeys = get_protein_values(peptides_txt_mapped_log2,intensity_columns) | 91 protein_intensities_tukeys = get_protein_values(peptides_txt_mapped_log2,intensity_columns) |