Mercurial > repos > vijay > pancancer_classifier
changeset 0:20b049f5b3f5 draft default tip
"planemo upload for repository http://github.com/nvk747/papaa/galaxy/ commit 954b283ef7f82f59f55476a4b3a230d655187ac1"
| author | vijay |
|---|---|
| date | Wed, 16 Dec 2020 23:32:43 +0000 |
| parents | |
| children | |
| files | macros.xml pancancer_classifier.xml |
| diffstat | 2 files changed, 436 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/macros.xml Wed Dec 16 23:32:43 2020 +0000 @@ -0,0 +1,229 @@ +<macros> + <token name="@VERSION@">0.1.9</token> + + <xml name="requirements"> + <requirements> + <requirement type="package" version="@VERSION@">papaa</requirement> + </requirements> + </xml> + + <xml name="stdio"> + <stdio> + <exit_code level="fatal" range="1:"/> + </stdio> + </xml> + + <xml name="inputs_basic"> + <param argument="--x_matrix" label="Filename of features to use in model" name="x_matrix" optional="true" type="data" format="tabular" help="data/pancan_rnaseq_freeze.tsv.gz"/> + <yield/> + <param argument="--filename_mut" label="Filename mutations" name="filename_mut" optional="true" type="data" format="tabular" help="data/pancan_mutation_freeze.tsv.gz"/> + <param argument="--filename_mut_burden" label="Filename of mutation burden" name="filename_mut_burden" optional="true" type="data" format="tabular" help="data/mutation_burden_freeze.tsv"/> + <param argument="--filename_sample" label="Filename of sample" name="filename_sample" optional="true" type="data" format="tabular" help="data/sample_freeze.tsv"/> + </xml> + + <token name="@INPUTS_BASIC@"><![CDATA[ + #if $x_matrix and $x_matrix is not None: + --x_matrix '$x_matrix' + #end if + #if $filename_mut and $filename_mut is not None: + --filename_mut '$filename_mut' + #end if + #if $filename_mut_burden and $filename_mut_burden is not None: + --filename_mut_burden '$filename_mut_burden' + #end if + #if $filename_sample and $filename_sample is not None: + --filename_sample '$filename_sample' + #end if + ]]> + </token> + + <xml name="inputs_genes_diseases"> + <param argument="--genes" label="Comma separated string of HUGO gene symbols" name="genes" optional="False" type="text" value="ERBB2,PIK3CA,KRAS,AKT1"/> + <param argument="--diseases" label="Comma sep string of TCGA disease acronyms. If no arguments are passed, filtering will default to options given in --filter_count and --filter_prop." name="diseases" optional="true" type="text" value="BLCA,BRCA,CESC,COAD,ESCA,LUAD,LUSC,OV,PRAD,READ,STAD,UCEC,UCS"/> + </xml> + + <token name="@INPUTS_GENES_DISEASES@"><![CDATA[ + #if $genes and $genes is not None: + --genes '$genes' + #end if + #if $diseases and str($diseases) != '': + --diseases '$diseases' + #end if + ]]> + </token> + + + <xml name="input_filename_raw_mut"> + <param argument="--filename_raw_mut" label="Filename of raw mut MAF" name="filename_raw_mut" optional="true" type="data" format="tabular" help="data/raw/mc3.v0.2.8.PUBLIC.maf"/> + </xml> + + <token name="@INPUT_FILENAME_RAW_MUT@"><![CDATA[ + #if $filename_raw_mut and $filename_raw_mut is not None: + --filename_raw_mut '$filename_raw_mut' + #end if + ]]> + </token> + + + <xml name="input_filename_burden"> + <param argument="--filename_burden" label="Burden file" name="filename_burden" optional="true" type="data" format="tabular" help="data/seg_based_scores.tsv"/> + </xml> + + <token name="@INPUT_FILENAME_BURDEN@"><![CDATA[ + #if $filename_burden and $filename_burden is not None: + --filename_burden '$filename_burden' + #end if + ]]> + </token> + + <xml name="input_filename_snaptron_samples"> + <param argument="--sample_file" label="SNAPTRON samples" name="sample_file" optional="true" type="data" format="tabular" help="scripts/snaptron/samples.tsv.gz"/> + </xml> + + <token name="@INPUT_FILENAME_SNAPTRON_SAMPLES@"><![CDATA[ + #if $sample_file and $sample_file is not None: + --sample_file '$sample_file' + #end if + ]]> + </token> + + + <xml name="input_filename_snaptron_junctions"> + <param argument="--junction_file" label="SNAPTRON junctions" name="junction_file" optional="true" type="data" format="tabular" help="scripts/snaptron/tp53_junctions.txt.gz"/> + </xml> + + <token name="@INPUT_FILENAME_SNAPTRON_JUNCTIONS@"><![CDATA[ + #if $junction_file and $junction_file is not None: + --junction_file '$junction_file' + #end if + ]]> + </token> + + <xml name="inputs_copy_number_file"> + <param argument="--filename_copy_loss" label="File with Copy number loss" name="filename_copy_loss" optional="true" type="data" format="tabular" help="data/copy_number_loss_status.tsv.gz"/> + <param argument="--filename_copy_gain" label="File with Copy number gain" name="filename_copy_gain" optional="true" type="data" format="tabular" help="data/copy_number_gain_status.tsv.gz"/> + </xml> + + <xml name="inputs_copy_number_class_file"> + <expand macro="inputs_copy_number_file" /> + <param argument="--filename_cancer_gene_classification" label="File with cancer gene classification table" name="filename_cancer_gene_classification" optional="true" type="data" format="tabular" help="data/cosmic_cancer_classification.tsv"/> + </xml> + + <xml name="inputs_copy_number_file_conditional"> + <conditional name="copy_number_conditional"> + <param argument="--copy_number" checked="false" label="Supplement Y matrix with copy number events" name="copy_number" type="boolean" truevalue="--copy_number" falsevalue=""/> + <when value="--copy_number"> + <expand macro="inputs_copy_number_file" /> + </when> + <when value=""/> + </conditional> + </xml> + + <xml name="inputs_copy_number_class_file_conditional"> + <conditional name="copy_number_conditional"> + <param argument="--copy_number" checked="false" label="Supplement Y matrix with copy number events" name="copy_number" type="boolean" truevalue="--copy_number" falsevalue=""/> + <when value="--copy_number"> + <expand macro="inputs_copy_number_class_file" /> + </when> + <when value=""/> + </conditional> + </xml> + + <token name="@INPUTS_COPY_NUMBER_FILE_PREFIX@"><![CDATA[ + #set $copy_number_conditional = type('',(object,),{'filename_copy_loss':$filename_copy_loss,'filename_copy_gain':$filename_copy_gain})() + ]]> + </token> + + <token name="@INPUTS_COPY_NUMBER_FILE@"><![CDATA[ + #if $copy_number_conditional.filename_copy_loss and $copy_number_conditional.filename_copy_loss is not None: + --filename_copy_loss '$copy_number_conditional.filename_copy_loss' + #end if + #if $copy_number_conditional.filename_copy_gain and $copy_number_conditional.filename_copy_gain is not None: + --filename_copy_gain '$copy_number_conditional.filename_copy_gain' + #end if + ]]> + </token> + + <token name="@INPUTS_COPY_NUMBER_FILE_CONDITIONAL@"><![CDATA[ + #if $copy_number_conditional.copy_number + $copy_number_conditional.copy_number + #if $copy_number_conditional.filename_copy_loss and $copy_number_conditional.filename_copy_loss is not None: + --filename_copy_loss '$copy_number_conditional.filename_copy_loss' + #end if + #if $copy_number_conditional.filename_copy_gain and $copy_number_conditional.filename_copy_gain is not None: + --filename_copy_gain '$copy_number_conditional.filename_copy_gain' + #end if + ##@INPUTS_COPY_NUMBER_FILE@ + #end if + ]]> + </token> + + <token name="@INPUTS_COPY_NUMBER_CLASS_FILE_PREFIX@"><![CDATA[ + #set $copy_number_conditional = type('',(object,),{'filename_copy_loss':$filename_copy_loss,'filename_copy_gain':$filename_copy_gain,'filename_cancer_gene_classification':$filename_cancer_gene_classification})() + ]]> + </token> + + <token name="@INPUTS_COPY_NUMBER_CLASS_FILE@"><![CDATA[ + #if $copy_number_conditional.filename_copy_loss and $copy_number_conditional.filename_copy_loss is not None: + --filename_copy_loss '$copy_number_conditional.filename_copy_loss' + #end if + #if $copy_number_conditional.filename_copy_gain and $copy_number_conditional.filename_copy_gain is not None: + --filename_copy_gain '$copy_number_conditional.filename_copy_gain' + #end if + #if $copy_number_conditional.filename_cancer_gene_classification and $copy_number_conditional.filename_cancer_gene_classification is not None: + --filename_cancer_gene_classification '$copy_number_conditional.filename_cancer_gene_classification' + #end if + ]]> + </token> + + <token name="@INPUTS_COPY_NUMBER_CLASS_FILE_CONDITIONAL@"><![CDATA[ + #if $copy_number_conditional.copy_number + $copy_number_conditional.copy_number + ## + #if $copy_number_conditional.filename_copy_loss and $copy_number_conditional.filename_copy_loss is not None: + --filename_copy_loss '$copy_number_conditional.filename_copy_loss' + #end if + #if $copy_number_conditional.filename_copy_gain and $copy_number_conditional.filename_copy_gain is not None: + --filename_copy_gain '$copy_number_conditional.filename_copy_gain' + #end if + #if $copy_number_conditional.filename_cancer_gene_classification and $copy_number_conditional.filename_cancer_gene_classification is not None: + --filename_cancer_gene_classification '$copy_number_conditional.filename_cancer_gene_classification' + #end if + ##@INPUTS_COPY_NUMBER_CLASS_FILE@ + #end if + ]]> + </token> + + <xml name="input_remove_hyper"> + <param argument="--remove_hyper" checked="false" label="Remove hypermutated samples" name="remove_hyper" type="boolean" truevalue="--remove_hyper" falsevalue=""/> + </xml> + <token name="@INPUT_REMOVE_HYPER@"><![CDATA[$remove_hyper]]></token> + + <xml name="input_alphas"> + <param argument="--alphas" label="the alphas for parameter sweep" name="alphas" optional="true" type="text" value="0.1,0.13,0.15,0.18,0.2,0.3,0.4,0.6,0.7"/> + </xml> + + <token name="@INPUTS_ALPHAS@"><![CDATA[ + #if $alphas and $alphas is not None: + --alphas '$alphas' + #end if + ]]> + </token> + + <xml name="input_l1_ratios"> + <param argument="--l1_ratios" label="the l1 ratios for parameter sweep" name="l1_ratios" optional="true" type="text" value="0.1,0.125,0.15,0.2,0.25,0.3,0.35"/> + </xml> + + <token name="@INPUTS_L1_RATIOS@"><![CDATA[ + #if $l1_ratios and $l1_ratios is not None: + --l1_ratios '$l1_ratios' + #end if + ]]> + </token> + <xml name="citations"> + <citations> + <yield /> + </citations> + </xml> +</macros> +
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/pancancer_classifier.xml Wed Dec 16 23:32:43 2020 +0000 @@ -0,0 +1,207 @@ +<tool id="pancancer_classifier" name="PAPAA: PanCancer classifier" version="@VERSION@" python_template_version="3.6"> + <description>classifier for pathway aberrant activity</description> + <macros> + <import>macros.xml</import> + </macros> + <expand macro="requirements"/> + <expand macro="stdio"/> + <version_command> + <![CDATA['papaa_pancancer_classifier.py' --version]]></version_command> + <command detect_errors="exit_code"><![CDATA[ + papaa_pancancer_classifier.py + #if $folds and $folds is not None: + --folds '$folds' + #end if + #if $seed and $seed is not None: + --seed '$seed' + #end if + $drop + #if $filter_count and $filter_count is not None: + --filter_count '$filter_count' + #end if + #if $filter_prop and $filter_prop is not None: + --filter_prop '$filter_prop' + #end if + #if $num_features and $num_features is not None: + --num_features '$num_features' + #end if + @INPUTS_ALPHAS@ + @INPUTS_L1_RATIOS@ + #if $alt_genes and str($alt_genes) != '': + --alt_genes '$alt_genes' + #end if + #if $alt_diseases and str($alt_diseases) != '': + --alt_diseases '$alt_diseases' + #end if + #if $alt_filter_count and $alt_filter_count is not None: + --alt_filter_count '$alt_filter_count' + #end if + #if $alt_filter_prop and $alt_filter_prop is not None: + --alt_filter_prop '$alt_filter_prop' + #end if + --classifier_results 'classifier' + @INPUT_REMOVE_HYPER@ + $keep_intermediate + @INPUTS_BASIC@ + @INPUTS_GENES_DISEASES@ + $drop + $shuffled + $shuffled_before_training + $no_mutation + #if str($drop_x_genes) + --drop_x_genes '$drop_x_genes' + #end if + $drop_expression + $drop_covariates + @INPUTS_COPY_NUMBER_CLASS_FILE_CONDITIONAL@ + > '${log}' + ]]> + </command> + <inputs> + <expand macro="inputs_basic" /> + <expand macro="inputs_genes_diseases" /> + <param argument="--seed" label="option to set seed" name="seed" optional="true" type="integer" value="1234"/> + <param argument="--folds" label="Number of cross validation folds to perform" name="folds" optional="true" type="integer" value="5"/> + <param argument="--drop" checked="false" label="Decision to drop input genes from X matrix" name="drop" type="boolean" truevalue="--drop" falsevalue=""/> + <expand macro="inputs_copy_number_class_file_conditional" /> + <param argument="--filter_count" label="Min number of mutations in diseases to include" name="filter_count" optional="true" type="integer" value="15"/> + <param argument="--filter_prop" label="Min proportion of positives to include disease" name="filter_prop" optional="true" type="float" value="0.05"/> + <param argument="--num_features" label="Number of MAD genes to include in classifier" name="num_features" optional="true" type="integer" value="8000"/> + <expand macro="input_alphas" /> + <expand macro="input_l1_ratios" /> + <param argument="--alt_genes" label="alternative genes to test performance" name="alt_genes" optional="true" type="text" /> + <param argument="--alt_diseases" label="The alternative diseases to test performance" name="alt_diseases" optional="true" type="text" /> + <param argument="--alt_filter_count" label="Min number of mutations in disease to include in alternate" name="alt_filter_count" optional="true" type="integer" value="15"/> + <param argument="--alt_filter_prop" label="Min proportion of positives to include disease in alternate" name="alt_filter_prop" optional="true" type="float" value="0.05"/> + <expand macro="input_remove_hyper" /> + <param argument="--keep_intermediate" checked="false" label="Keep intermediate ROC values for plotting" name="keep_intermediate" type="boolean" truevalue="--keep_intermediate" falsevalue=""/> + <param argument="--shuffled" checked="false" label="Shuffle the input gene exprs matrix alongside" name="shuffled" type="boolean" truevalue="--shuffled" falsevalue=""/> + <param argument="--shuffled_before_training" checked="false" label="Shuffle the gene exprs matrix before training" name="shuffled_before_training" type="boolean" truevalue="--shuffled_before_training" falsevalue=""/> + <param argument="--no_mutation" checked="false" label="Remove mutation data from y matrix" name="no_mutation" type="boolean" truevalue="--no_mutation" falsevalue=""/> + <param argument="--drop_x_genes" label="Comma separated list of genes to be dropped from X matrix" name="drop_x_genes" optional="true" type="text" value=""/> + <param argument="--drop_expression" checked="false" label="Decision to drop gene expression values from X" name="drop_expression" type="boolean" truevalue="--drop_expression" falsevalue=""/> + <param argument="--drop_covariates" checked="false" label="Decision to drop covariate information from X" name="drop_covariates" type="boolean" truevalue="--drop_covariates" falsevalue=""/> + </inputs> + <outputs> + <data format="txt" name="log" label="${tool.name} on ${on_string} (Log)" /> + <data format="tabular" name="alt_gene_alt_disease_summary" label="${tool.name} on ${on_string} (alt_gene_alt_disease_summary.tsv)" from_work_dir="classifier/alt_gene_alt_disease_summary.tsv"/> + <data format="csv" name="alt_summary_counts" label="${tool.name} on ${on_string} (alt_summary_counts.csv" from_work_dir="classifier/alt_summary_counts.csv"/> + <data format="tabular" name="classifier_coefficients" label="${tool.name} on ${on_string} (classifier_coefficients.tsv)" from_work_dir="classifier/classifier_coefficients.tsv"/> + <data format="txt" name="classifier_summary" label="${tool.name} on ${on_string} (classifier_summary.txt)" from_work_dir="classifier/classifier_summary.txt"/> + <data format="tabular" name="pancan_roc_results" label="${tool.name} on ${on_string} (pancan_roc_results.tsv)" from_work_dir="classifier/pancan_roc_results.tsv"/> + <data format="csv" name="summary_counts" label="${tool.name} on ${on_string} (summary_counts.csv)" from_work_dir="classifier/summary_counts.csv"/> + <data format="pdf" name="cv_heatmap" label="${tool.name} on ${on_string} (cv_heatmap.pdf)" from_work_dir="classifier/cv_heatmap.pdf"/> + <collection name="disease_figures" type="list:list" label="Disease classifier figures"> + <discover_datasets pattern="classifier_(?P<identifier_1>.+)__pred_(?P<identifier_0>.+)\.pdf" format="pdf" directory="classifier/disease" visible="false" /> + </collection> + <data format="pdf" name="all_disease_pr" label="${tool.name} on ${on_string} (all_disease_pr.pdf)" from_work_dir="classifier/all_disease_pr.pdf"/> + <data format="pdf" name="all_disease_roc" label="${tool.name} on ${on_string} (all_disease_roc.pdf)" from_work_dir="classifier/all_disease_roc.pdf"/> + <data format="pdf" name="alt_gene_alt_disease_aupr_bar" label="${tool.name} on ${on_string} (alt_gene_alt_disease_aupr_bar.pdf)" from_work_dir="classifier/alt_gene_alt_disease_aupr_bar.pdf"/> + <data format="pdf" name="alt_gene_alt_disease_auroc_bar" label="${tool.name} on ${on_string} (alt_gene_alt_disease_auroc_bar.pdf)" from_work_dir="classifier/alt_gene_alt_disease_auroc_bar.pdf"/> + <data format="pdf" name="disease_aupr" label="${tool.name} on ${on_string} (disease_aupr.pdf)" from_work_dir="classifier/disease_aupr.pdf"/> + <data format="pdf" name="disease_auroc" label="${tool.name} on ${on_string} (disease_auroc.pdf)" from_work_dir="classifier/disease_auroc.pdf"/> + </outputs> + <tests> + <test> + <param name="genes" value="ERBB2,PIK3CA,KRAS,AKT1"/> + <param name="diseases" value="GBM"/> + <param name="x_matrix" value="pancan_rnaseq_freeze_t1p.tsv.gz" ftype="tabular"/> + <param name="filename_mut" value="pancan_mutation_freeze_t1p.tsv.gz" ftype="tabular"/> + <param name="filename_mut_burden" value="mutation_burden_freeze.tsv" ftype="tabular"/> + <param name="filename_sample" value="sample_freeze.tsv" ftype="tabular"/> + <param name="seed" value="1234"/> + <param name="folds" value="5"/> + <param name="drop" value="true"/> + <param name="copy_number" value="true"/> + <param name="filename_copy_loss" value="copy_number_loss_status_t10p.tsv.gz" ftype="tabular"/> + <param name="filename_copy_gain" value="copy_number_gain_status_t10p.tsv.gz" ftype="tabular"/> + <param name="filename_cancer_gene_classification" value="cosmic_cancer_classification.tsv" ftype="tabular"/> + <param name="filter_count" value="15"/> + <param name="filter_prop" value="0.05"/> + <param name="num_features" value="8000"/> + <param name="alphas" value="0.1,0.13,0.15,0.18,0.2,0.3,0.4,0.6,0.7"/> + <param name="l1_ratios" value="0.1,0.125,0.15,0.2,0.25,0.3,0.35"/> + <param name="alt_genes" value="PTEN,PIK3R1,STK11"/> + <param name="alt_diseases" value="GBM"/> + <param name="alt_filter_count" value="15"/> + <param name="alt_filter_prop" value="0.05"/> + <param name="remove_hyper" value="true"/> + <param name="keep_intermediate" value="true"/> + <param name="shuffled" value="true"/> + <param name="shuffled_before_training" value="false"/> + <param name="no_mutation" value="false"/> + <param name="drop_expression" value="false"/> + <param name="drop_covariates" value="false"/> + <output name="log" file="Log.txt"/> + <output name="alt_gene_alt_disease_summary" file="alt_gene_alt_disease_summary.tsv"/> + <output name="alt_summary_counts" file="alt_summary_counts.csv"/> + <output name="classifier_coefficients" file="classifier_coefficients.tsv"/> + <output name="classifier_summary" file="classifier_summary.txt"/> + <output name="pancan_roc_results"> + <assert_contents> + <has_line line="	fpr	tpr	threshold	train_type	disease" /> + <has_n_columns n="6" /> + <has_n_lines n="253" /> + </assert_contents> + </output> + <output name="summary_counts" file="summary_counts.csv"/> + <output name="cv_heatmap" file="cv_heatmap.pdf" compare="sim_size" delta="50"/> + <output name="all_disease_pr" file="all_disease_pr.pdf" compare="sim_size" delta="50" /> + <output name="all_disease_roc" file="all_disease_roc.pdf" compare="sim_size" delta="50"/> + <output name="alt_gene_alt_disease_aupr_bar" file="alt_gene_alt_disease_aupr_bar.pdf" compare="sim_size" delta="50"/> + <output name="alt_gene_alt_disease_auroc_bar" file="alt_gene_alt_disease_auroc_bar.pdf" compare="sim_size" delta="50"/> + <output name="disease_aupr" file="disease_aupr.pdf" compare="sim_size" delta="50"/> + <output name="disease_auroc" file="disease_auroc.pdf" compare="sim_size" delta="50"/> + </test> + </tests> + <help><![CDATA[ + + **Pancancer_Aberrant_Pathway_Activity_Analysis scripts/papaa_pancancer_classifier.py:** + + **Inputs:** + --genes comma separated string of HUGO symbols for target genes or targenes_list.csv file + --diseases comma separated string of disease types/TCGA acronyms for classifier + default: Auto (will pick diseases from filter args) + --folds number of cross validation folds + default: 5 + --seed value specifies the initial value of the random number seed + default: 1234 + --drop drop the input genes from the X matrix + default: False if flag omitted + --copy_number optional flag to supplement copy number to define Y + default: False if flag omitted + --filter_count int of low count of mutation to include disease + default: 15 + --filter_prop float of low proportion of mutated samples per disease + default: 0.05 + --num_features int of number of genes to include in classifier + default: 8000 + --alphas comma separated string of alphas to test in pipeline default: '0.1,0.15,0.2,0.5,0.8,1' + --l1_ratios comma separated string of l1 parameters to test + default: '0,0.1,0.15,0.18,0.2,0.3' + --alt_genes comma separated string of alternative genes to test + default: None + --alt_diseases comma separated string of alternative diseases to test + default: Auto + --alt_filter_count int of low count of mutations to include alt_diseases + default: 15 + --alt_filter_prop float of low proportion of mutated samples alt_disease + default: 0.05 + --classifier_results string of the location to save the classifier results/figures + default: Auto + --remove_hyper store_true: remove hypermutated samples + default: False if flag omitted + --keep_intermediate store_true: keep intermediate roc curve items + default: False if flag omitted + --x_matrix string of which feature matrix to use + default: raw + --remove_hyper store_true: remove hypermutated samples + default: False if flag omitted + --keep_intermediate store_true: keep intermediate roc curve items + default: False if flag omitted + + **Outputs:** + ROC curves, AUROC across diseases, and classifier coefficients ]]> + </help> + <expand macro="citations" /> +</tool>