comparison COBRAxy/flux_to_map.py @ 302:753347af8bc9 draft

Uploaded

301:5a595a737220

class ArrowColor(Enum):
    """
    Encodes possible arrow colors based on their meaning in the enrichment process.
    """
    Invalid       = "#BEBEBE" # gray, fold-change under treshold
    Transparent   = "#ffffff00" # white, not significant p-value
    UpRegulated   = "#ecac68" # red, up-regulated reaction
    DownRegulated = "#6495ed" # blue, down-regulated reaction

    UpRegulatedInv = "#FF0000"
    # ^^^ different shade of red (actually orange), up-regulated net value for a reversible reaction with

    match ARGS.test:
        case "ks":
            # Perform Kolmogorov-Smirnov test
            _, p_value = st.ks_2samp(dataset1Data, dataset2Data)
        case "ttest_p":
            # Perform t-test for paired samples
            _, p_value = st.ttest_rel(dataset1Data, dataset2Data)
        case "ttest_ind":
            # Perform t-test for independent samples
            _, p_value = st.ttest_ind(dataset1Data, dataset2Data)
        case "wilcoxon":
            # Perform Wilcoxon signed-rank test
            _, p_value = st.wilcoxon(dataset1Data, dataset2Data)
        case "mw":
            # Perform Mann-Whitney U test
            _, p_value = st.mannwhitneyu(dataset1Data, dataset2Data)

        except (TypeError, ZeroDivisionError): continue

    # Apply multiple testing correction if set by the user
    if ARGS.adjusted:
        
        # Retrive the p-values from the comparisonResult dictionary
        reactIds = list(comparisonResult.keys())
        pValues = [comparisonResult[reactId][0] for reactId in reactIds]
        
        # Apply the Benjamini-Hochberg correction and update
        adjustedPValues = st.multipletests(pValues, method="fdr_bh")[1]
        for i, reactId in enumerate(reactIds):
            comparisonResult[reactId][0] = adjustedPValues[i]
    
    return comparisonResult, max_z_score

def computeEnrichment(class_pat :Dict[str, List[List[float]]], ids :List[str]) -> List[Tuple[str, str, dict, float]]:
    """

302:753347af8bc9

class ArrowColor(Enum):
    """
    Encodes possible arrow colors based on their meaning in the enrichment process.
    """
    Invalid       = "#BEBEBE" # gray, fold-change under treshold or not significant p-value
    Transparent   = "#ffffff00" # transparent, to make some arrow segments disappear
    UpRegulated   = "#ecac68" # red, up-regulated reaction
    DownRegulated = "#6495ed" # blue, down-regulated reaction

    UpRegulatedInv = "#FF0000"
    # ^^^ different shade of red (actually orange), up-regulated net value for a reversible reaction with

    match ARGS.test:
        case "ks":
            # Perform Kolmogorov-Smirnov test
            _, p_value = st.ks_2samp(dataset1Data, dataset2Data)
        case "ttest_p":
            # Datasets should have same size
            if len(dataset1Data) != len(dataset2Data):
                raise ValueError("Datasets must have the same size for paired t-test.")
            # Perform t-test for paired samples
            _, p_value = st.ttest_rel(dataset1Data, dataset2Data)
        case "ttest_ind":
            # Perform t-test for independent samples
            _, p_value = st.ttest_ind(dataset1Data, dataset2Data)
        case "wilcoxon":
            # Datasets should have same size
            if len(dataset1Data) != len(dataset2Data):
                raise ValueError("Datasets must have the same size for Wilcoxon signed-rank test.")
            # Perform Wilcoxon signed-rank test
            _, p_value = st.wilcoxon(dataset1Data, dataset2Data)
        case "mw":
            # Perform Mann-Whitney U test
            _, p_value = st.mannwhitneyu(dataset1Data, dataset2Data)

        except (TypeError, ZeroDivisionError): continue

    # Apply multiple testing correction if set by the user
    if ARGS.adjusted:
        
        # Retrieve the p-values from the comparisonResult dictionary, they have to be different from NaN
        validPValues = [(reactId, result[0]) for reactId, result in comparisonResult.items() if not np.isnan(result[0])]

        if not validPValues:
            return comparisonResult, max_z_score
        
        # Unpack the valid p-values
        reactIds, pValues = zip(*validPValues)
        # Adjust the p-values using the Benjamini-Hochberg method
        adjustedPValues = st.false_discovery_control(pValues)
        # Update the comparisonResult dictionary with the adjusted p-values
        for reactId , adjustedPValue in zip(reactIds, adjustedPValues):
            comparisonResult[reactId][0] = adjustedPValue
    
    return comparisonResult, max_z_score

def computeEnrichment(class_pat :Dict[str, List[List[float]]], ids :List[str]) -> List[Tuple[str, str, dict, float]]:
    """