Log-Contrast Regression with TRAC (Taxonomic Information)

Log-Contrast Regression with TRAC (Taxonomic Information)#

This tutorial demonstrates how to leverage taxonomic hierarchical information in log-contrast regression to predict average soil temperature. TRAC (Taxonomic Regularized Adaptive Contrasts) improves feature selection by incorporating the taxonomic structure.

Overview#

TRAC regression incorporates taxonomic classifications and computes adaptive weights based on the taxonomic hierarchy. This approach can identify which taxonomic groups (not just individual taxa) are most predictive of your target variable, leading to more interpretable and potentially more robust predictions.

Step 1: Transform Features#

Apply CLR transformation to the count data:

qiime classo transform-features \
    --p-transformation clr \
    --p-coef 0.5 \
    --i-features data/atacama-counts.qza \
    --o-x data/xclr

Step 2: Add Taxonomic Information#

Incorporate taxonomic classifications and compute adaptive weights based on taxonomy:

qiime classo add-taxa \
    --i-features data/xclr.qza  \
    --i-taxa data/classification.qza \
    --o-x data/xtaxa \
    --o-aweights data/wtaxa

This step:

  • Uses taxonomic hierarchy to structure feature relationships

  • Computes adaptive weights based on taxonomic distances

  • Encourages selection of taxa that share taxonomic relatives

Step 3: Add Covariates#

Include environmental metadata with custom weights, combined with taxonomic features:

qiime classo add-covariates \
    --i-features data/xtaxa.qza \
    --i-weights data/wtaxa.qza \
    --m-covariates-file data/atacama-selected-covariates-veg.tsv \
    --p-to-add ph average-soil-relative-humidity elevation average-soil-temperature vegetation \
    --p-w-to-add 1. 0.1 0.1 0.1 1 \
    --o-new-features data/xcovariates_trac \
    --o-new-c data/ccovariates_trac \
    --o-new-w data/wcovariates_trac

Step 4: Split Data#

Create training and test sets for regression analysis:

qiime sample-classifier split-table \
    --i-table data/xcovariates_trac.qza \
    --m-metadata-file data/atacama-selected-covariates-veg.tsv \
    --m-metadata-column average-soil-temperature \
    --p-test-size 0.2 \
    --p-random-state 42 \
    --p-stratify False \
    --o-training-table data/regress-xtraining_trac.qza \
    --o-test-table data/regress-xtest_trac.qza \
    --o-training-targets data/regress-training-targets_trac.qza \
    --o-test-targets data/regress-test-targets_trac.qza

Step 5: Train the Regression Model#

Use log-contrast regression with taxonomic information and stability selection:

qiime classo regress \
    --i-features data/regress-xtraining_trac.qza \
    --i-c data/ccovariates_trac.qza \
    --i-weights data/wcovariates_trac.qza \
    --m-y-file data/atacama-selected-covariates-veg.tsv \
    --m-y-column average-soil-temperature \
    --p-concomitant False \
    --p-stabsel \
    --p-cv \
    --p-path \
    --p-lamfixed \
    --p-stabsel-threshold 0.5 \
    --p-cv-seed 1 \
    --p-no-cv-one-se \
    --o-result data/regresstaxa_trac.qza

Key parameters:

  • --p-stabsel: Enable stability selection for robust feature selection

  • --p-stabsel-threshold 0.5: Features selected in >50% of subsamples

  • --p-cv: Use cross-validation for model selection

  • --p-concomitant False: Use standard formulation

Step 6: Make Predictions#

Apply the trained model to test data:

qiime classo predict \
    --i-features data/regress-xtest_trac.qza \
    --i-problem data/regresstaxa_trac.qza \
    --o-predictions data/regress-predictions_trac.qza

Step 7: Visualize Results#

Generate a comprehensive summary including taxonomic information:

qiime classo summarize \
    --i-problem data/regresstaxa_trac.qza \
    --i-taxa data/classification.qza \
    --i-predictions data/regress-predictions_trac.qza \
    --o-visualization data/regresstaxa_R1_trac.qzv

Output visualization: The .qzv file contains:

  • Selected taxa with taxonomic context

  • Model performance metrics (R², RMSE, etc.)

  • Taxonomic group importances

  • Cross-validation curves

  • Prediction accuracy on test data

View the results at QIIME 2 View.

Comparing Log-Contrast vs TRAC#

Log-Contrast:

  • Treats each taxon independently

  • Faster computation

  • Useful for exploratory analysis

TRAC:

  • Leverages taxonomic relationships

  • More interpretable results

  • Better feature selection for hierarchically structured data

  • Slightly increased computation time

Contents