Hello Serge,

I tried to reproduce your results without success. My search using the rentrez package returned 29044 hits, while the web (queried as: https://www.ncbi.nlm.nih.gov/nuccore?term=Argasidae%5BOrganism%5D) just returned the same number of hits. I also tried other Taxa names and both numbers agreeded as intended. I think it could be a temporal issue related to an actualization of the database by deleting obsolete entries.

To look for discrepancies I would just get the webpages concerning all the Taxa you are interested in (as in the link I provided) in R and do some scraping with rvest to programmatically check for this "Items: 1 to 20 of 29044" inside the html file. There you could do the same query nearly at the same time both with rentrez and scraping with rvest. Example code below (you can just made a function out of it and feed it the Taxa name you want)

> library(rvest)

> web unlist(strsplit(as.vector(web %>% html_nodes('h2') %>% html_text())[2], ' '))[6]

This returns "29044"

Hello, Leopoldo.

Many thanks for your reply. I think it provided some clues to the discrepancy because your query at NCBI contained the search term [Organism] whereas mine did not. So, the results included some Rickettsia sequences (a bacterium found in Argasidae) as well as some unrelated ticks (e.g. Rhipicephalus microplus). I think the answer to my question is kind of obvious although I did not expect that "search all fields" should be a default when querying GenBank...

Best regards,

Serhii

Well, I am glad you noticed then. That surely explains it all.