-
-
-Code
-#all spacers from CCtyper were combined into a tsv file as FASTA format
-all_spacers <- read_tsv(here("joining CCtyper/all_spacers.tsv"), show_col_types=FALSE)
-names(all_spacers) <- c("Sample_accession", "Spacer_sequence")
-
-#contig index are extracted from sample accession and put into their own column for pattern analysis
-all_spacers[3] <- str_extract(all_spacers$Sample_accession, "\\d*:\\d*")
-all_spacers <- separate(all_spacers, 3, into = c("Array_index", "Spacer_index"))
-all_spacers["Contig_accession"] <- str_extract(all_spacers$Sample_accession, "contig\\d*")
-
-#clean up sample accession
-all_spacers$Sample_accession <- str_extract(all_spacers$Sample_accession, "(?<=>)\\w*")
-
-#these spacers also include repeats and can be orphans or connected to a cas operon, this is added to the all spacer frame.
-orphan_crispr <- read_tsv(here("joining CCtyper/crisprs_orphan-concatenated.tab"), show_col_types = F)[,c(3, 6)] %>% mutate("orphan?" = TRUE) %>% separate(col = 1, sep = "\\.|\\_", into = c("Sample_accession", "Contig_accession", "Array_index"))
-orphan_crispr$Contig_accession <- str_extract(orphan_crispr$Contig_accession, "^\\w*")
-
-
-crispr_cas <- read_tsv(here("joining CCtyper/crisprs_near_cas-concatenated.tab"), show_col_types = F)[,c(3, 6)] %>% mutate("orphan?" = FALSE) %>% separate(col = 1, sep = "\\.|\\_", into = c("Sample_accession", "Contig_accession", "Array_index"))
-crispr_cas$Contig_accession <- str_extract(crispr_cas$Contig_accession, "^\\w*")
-
-all_spacers <- left_join(all_spacers, bind_rows(orphan_crispr, crispr_cas))
-
-#ENA metadata is downloaded from ATB and was filtered for original analysis in descriptive_statistics
-colnames <- read_table(here("metadata project/colnames_metadata_filtered.txt"), col_names = F, show_col_types = F)
-metadata_source <-
- read_tsv(here("metadata project/ena_metadata.20240801.selection-only_Campylobacter.tsv"),
- col_names = colnames$X2, show_col_types = FALSE)
-
-#further filtering metadata as not all are needed for this analysis
-metadata <- select(metadata_source, c(sample_accession, scientific_name, isolation_source, host))
-
-#in the metadata there is a distinction between unclear data and data that is simply missing/unspecified. missing data is turned into NA and data that is present in isolation source but not in host is copied to host.
-missingno <- metadata %>%
- mutate(across(c(isolation_source, host), ~str_to_lower(.x)))%>%
- replace_with_na(replace = list(isolation_source = c(
- "missing", "other", "not collected", "no source specified", "mising"
- ), host = c(
- "missing", "other", "not collected", "no source specified", "mising"
- ))) %>% mutate(host = ifelse(is.na(host), isolation_source, host))
-
-#There are also bacteria that are incorrectly noted in the ENA metadata compared to the ATB designation of species
-simplified_species <- read_delim(
- file = here("metadata project/sylph.tsv.gz"),
- delim = "\t",
- show_col_types = F) %>%
- select(Sample, Species)
-
-simplified_species <- simplified_species %>% filter(str_detect(Species,"Campylobacter_D (jejuni|coli)")) %>% mutate(Species = str_replace_all(Species, "_(A|B|C|D)", ""))
-
-merged_metadata <- left_join(missingno, simplified_species, by = c("sample_accession" = "Sample")) %>% distinct(sample_accession, .keep_all = TRUE)
