Add R API & usage example

[yannickwurm]

Subtasks?

1. Make a mini wrapper / R library for doing those commands? https://sequenceserver.com/doc/api/
2. And some examples of how to take the extended tab-delimited results, and filter them.
3. Add mechanism for it to work through SequenceServer Cloud authentication mechanism (magiclinks…)
4. (no need to make it possible to programatically add new databases at this stage).
5. Create documentation / blog post

We have an existing way of interrogating from shell (https://sequenceserver.com/doc/api/).

And an R example online: https://sequenceserver.com/blog/disentangling-homology-orthology-parology-and-similarity-with-BLAST

Here is some R code for getting the

# Loading data
blast_results_file_tsv = "sequenceserver-full_tsv_report.tsv" # although https example is better!
blast_results <- read.delim(blast_results_file_tsv, header = FALSE, comment.char = "#")
blast_results_header <- grep(pattern = "# Fields:", x = readLines(con = blast_results_file_tsv, n = 10), value = TRUE)[1]
blast_results_header <- sub(pattern = "# Fields: ", replacement = "", x = blast_results_header)
blast_results_colnames <- unlist(strsplit(x = blast_results_header, split = ", "))
blast_results_colnames <- gsub(pattern = " ", replacement = "_", x = blast_results_colnames)
colnames(blast_results) <- blast_results_colnames

# Example of pulling out top hit for each query
library(dplyr)
top_hits <- blast_results %>%
  arrange(query_id, desc(bit_score)) %>%
  group_by(query_id) %>%
  slice_head(n = 1) %>%
  arrange(desc(bit_score))