diff options
| -rw-r--r-- | wqflask/wqflask/correlation/rust_correlation.py | 39 |
1 files changed, 19 insertions, 20 deletions
diff --git a/wqflask/wqflask/correlation/rust_correlation.py b/wqflask/wqflask/correlation/rust_correlation.py index 4bd2dd9d..161215c5 100644 --- a/wqflask/wqflask/correlation/rust_correlation.py +++ b/wqflask/wqflask/correlation/rust_correlation.py @@ -13,6 +13,24 @@ from gn3.db_utils import database_connector +def compute_top_n_lit(corr_results, this_dataset, this_trait): + (this_trait_geneid, geneid_dict, species) = do_lit_correlation( + this_trait, this_dataset) + + geneid_dict = {trait_name: geneid for (trait_name, geneid) in geneid_dict.items() if + corr_results.get(trait_name)} + + conn = database_connector() + + with conn: + + correlation_results = compute_all_lit_correlation( + conn=conn, trait_lists=list(geneid_dict.items()), + species=species, gene_id=this_trait_geneid) + + return correlation_results + + def compute_top_n_tissue(this_dataset, this_trait, traits, method): @@ -56,29 +74,10 @@ def compute_correlation_rust(start_vars: dict, corr_type: str, lts = [key] + [str(x) for x in val] r = ",".join(lts) target_data.append(r) - # breakpoint() - results_k = run_correlation(target_data, ",".join( + results = run_correlation(target_data, ",".join( [str(x) for x in list(sample_data.values())]), method, ",") - tissue_top = compute_top_n_tissue( - this_dataset, this_trait, results_k, method) - - - lit_top = compute_top_n_lit(results_k,this_dataset,this_trait) - - - results = [] - - for (key,val) in results_k.items(): - if key in tissue_top: - results_k[key].update(tissue_top[key]) - - if key in lit_top: - results_k[key].update(lit_top[key]) - - results.append({key:results_k[key]}) - if corr_type == "tissue": |