Screens assemblies against a Mash database and predicts isolate source using a trained machine learning model mash pandas scikit-learn '${input_name}_mash.tabular' && ## 3. Conditional Logic: Check if Mash produced hits if [ \$(wc -l < '${input_name}_mash.tabular') -ge 2 ]; then python $__tool_directory__/predict_source.py -i '${input_name}_mash.tabular' -b '$model_selection.fields.path' -t $threshold -o '$prediction_output'; ## Optional: If the script ran but produced no results (e.g. filtered out) ## ensure we still provide the "No Prediction" fallback if [ ! -s '$prediction_output' ] || [ \$(wc -l < '$prediction_output') -le 1 ]; then echo -e "Run\tPredicted_Source\tConfidence_Score" > '$prediction_output'; echo -e "${input_name}\tNo Prediction\t0.0" >> '$prediction_output'; fi else ## Skip Python and create the fallback file immediately echo -e "Run\tPredicted_Source\tConfidence_Score" > '$prediction_output'; echo -e "${input_name}\tNo Prediction\t0.0" >> '$prediction_output'; fi ]]> @misc{strain_2026_plasmidtrakr, author = {Strain, Errol}, title = {PlasmidTrakr: A tool for predicting isolate source from plasmid profiles}, year = {2026}, publisher = {GitHub}, journal = {GitHub repository}, howpublished = {\url{https://github.com/estrain/plasmidtrakr}} }