Mercurial > repos > bgruening > simsearch
changeset 12:526f6e88fbe7
ChemicalToolBoX update.
| author | Bjoern Gruening <bjoern.gruening@gmail.com> |
|---|---|
| date | Fri, 31 May 2013 22:33:03 +0200 |
| parents | 2efd2a7c3015 |
| children | dbe9c82028d1 |
| files | simsearch.xml |
| diffstat | 1 files changed, 42 insertions(+), 7 deletions(-) [+] |
line wrap: on
line diff
--- a/simsearch.xml Mon May 27 17:00:09 2013 +0200 +++ b/simsearch.xml Fri May 31 22:33:03 2013 +0200 @@ -1,6 +1,5 @@ -<tool id="chemfp_simsearch" name="Similarity Search" version="0.1"> +<tool id="chemfp_simsearch" name="Similarity Search" version="0.1.1"> <description>of fingerprint data sets</description> - <!--<parallelism method="multi" split_inputs="query" split_mode="to_size" split_size="10000" shared_inputs="targets,fastsearch" merge_outputs="outfile"></parallelism>--> <requirements> <requirement type="package" version="1.1p1">chemfp</requirement> <requirement type="package" version="2.3.2">openbabel</requirement> @@ -8,7 +7,29 @@ <command> #if $method_opts.method_opts_selector == "chemfp": ## TODO: remove all comment lines, in muti mode many comment lines will be appear, also in the middle of the result file, remove them - simsearch -k $method_opts.k --threshold $method_opts.threshold --query-format fps --target-format fps -q "${method_opts.query}" "${method_opts.targets}" -o "${outfile}" + #if int($method_opts.knn) == 0: + #set $k = 'all' + #else: + #set $k = int($method_opts.knn) + #end if + simsearch + -k $k + --threshold $method_opts.threshold + --query-format fps + $method_opts.counts + --target-format fps + -o "${outfile}" + + ## build and search an in-memory data structure (faster for multiple queries) + --memory + + #if $method_opts.query_opts.query_opts_selector == "normal": + -q "${method_opts.query_opts.query}" + #else: + --NxN + #end if + + "${method_opts.query_opts.targets}" #else: ## OpenBabel needs the original molecule file (molecule.'fileformat') next to the fastsearch index (molecule.fs). We use a composite datatype to accomplish that. ## Furthermore OpenBabel is really picky with fileextensions. We need to specify every datatype. I did not find a solution to specify the query-filetype. @@ -30,10 +51,24 @@ <option value="obabel">OpenBabel Fastsearch Index</option> </param> <when value="chemfp"> - <param name="query" type='data' format="fps" label="query"/> - <param name="targets" type='data' format="fps" label="Target database"/> - <param name="k" type='text' label="k nearest neighbor" value='all'/> - <param name="threshold" type='float' label="threshold" value='0.7'/> + <conditional name="query_opts"> + <param name="query_opts_selector" type="select" label="Query Mode"> + <option value="normal">Query molecules are stores in a separate file</option> + <option value="nxn">Target molecules are also queries (NxN)</option> + </param> + <when value="normal"> + <param name='query' type='data' format="fps" label='Query molecules'/> + <param name='targets' type='data' format="fps" label='Target molecules'/> + </when> + <when value="nxn"> + <param name='targets' type='data' format="fps" label='Target moleculs'/> + </when> + </conditional> + <param name='knn' type='integer' value='0' label='select the k nearest neighbors' help='0 means all neighbors'> + <validator type="in_range" min="0" /> + </param> + <param name='threshold' type='float' label='threshold' value='0.7'/> + <param name="counts" type="boolean" label="report counts (-c)" truevalue="-c" falsevalue="" checked="false" /> </when> <when value="obabel"> <param name="query" type='data' format="smi,mol,sdf,inchi" label="query"/>