Filter returns wrong results

[fxstr]

Returns gram+ for gram-

[fxstr]

@Alpascal Should be fixed. Please provide some test cases to me (what to enter in the search field, what results to expect – especially with combinated search terms). Thanks a lot.

[pascalmfrey]

test cases: "gram+" should return bacterial species with the name containing "Staphylococcus" or "Streptococcus", and none of "Klebsiella", "Serratia", "Neisseria", "Pseudomonas" (and others). "gram-" should return the opposite (all "Klebsiella", none "Staphylococcus").

Combination of "gram+" and "coccus" should return species of "Staphylococcus" and "Streptococcus", and none of the ones above that are returned by "gram -".
Other combinations: gram+ & rod -> Clostridium; gram- & coccus -> Neisseria; gram- & rod -> Klebsiella, E. coli.

Does this help? Unfortunately I have very limited IT infrastructure here in Sri Lanka, but will do what's possible from here. Thanks for the work!

[fxstr]

gram+ looks good so far:

[fxstr]

gram-

[fxstr]

gram+ & coccus:

[fxstr]

gram- / rod

[fxstr]

Looks good, right? Thaks a lot for the cases, @Alpascal !

[pascalmfrey]

Unfortunately it's not correct yet, wrong results are returned in all, gram+, gram- and shape filters.
The following species must not be returned when filter gram+ is active:

When filter gram- ist active, the following must not be returned:

Could it be a problem with labelling in different language versions, i.e. when the filter term "rod" is entered, it should show all bacteria not only labelled as "rod" in the english language version but also "bacillus" as the "db entry term"?

The above list examples are from the old .xls file located here: infect/docs/Antibiotika Spektrum.xls

Species in the first list are all gram-, while species from the second are all gram+; however, there are further species in our database not in any of these lists that also have a gram property (but I think it's easiest if I look at those as soon as the filter works with the examples given above, as the follow the same structural rules)

[pascalmfrey]

As for the shapes:

• Acinetobacter, Aeromonas, Shigella, Stenotrophomonas, and Yersinia are not cocci, while both species of Enterococcus are gram+ cocci and not returned
• both Enterococcus and both Neisseria species are cocci, not rods, while Klebsiella, Pseudomonas and Proteus are both gram- rods but not returend

Thanks for looking into it!

[fxstr]

Coccus AND Gram+ return:

[fxstr]

@tobiaskneubuehler Will do a release, you'll be able to test online soon! Yeah.

[pascalmfrey]

Does the "coccus" and "gram+" also return the other species like
Streptococcus and Staphylococcus?
I will give it a thorough test as soon as it's online and my internet
connection is a bit faster than barely 28kbps...
On Feb 20, 2015 11:35 PM, "Felix Steiner" notifications@github.com wrote:

Coccus AND Gram+ return:
[image: screen shot 2015-02-20 at 19 01 06]
https://cloud.githubusercontent.com/assets/1211144/6306336/dd1ed354-b932-11e4-97fe-6675d8fc76bb.PNG

—
Reply to this email directly or view it on GitHub
#22 (comment).

[fxstr]

Here's the full list – and yes, it does. Hope that's the expected behaviour.

[pascalmfrey]

looks lovely! :-)
On Feb 21, 2015 3:05 PM, "Felix Steiner" notifications@github.com wrote:

Here's the full list – and yes, it does. Hope that's the expected
behaviour.

[image: screen shot 2015-02-21 at 10 33 04]
https://cloud.githubusercontent.com/assets/1211144/6313598/4e609a66-b9b5-11e4-9566-f08d9c80f9c8.png

—
Reply to this email directly or view it on GitHub
#22 (comment).