|News and Announcements »|
This script trains a supervised classifier using OTUs (or other continuous input sample x observation data) as predictors, and a mapping file column containing discrete values as the class labels.
It is recommended that you remove low-depth samples and rare OTUs before running this script. This can drastically reduce the run-time, and in many circumstances will not hurt performance. It is also recommended to perform rarefaction to control for sampling effort before running this script. For example, to rarefy at depth 200, then remove OTUs present in < 10 samples run:
For an overview of the application of supervised classification to microbiota, see PubMed ID 21039646.
This script also has the ability to collate the supervised learning results produced on an input directory. For example, in order to reduce any variation introduced through producing a rarefied OTU table, the user can run multiple_rarefactions_even_depth.py on the OTU table, and then pass that directory into supervised_learning.py. The user can then pass a -w collate_results filepath to produce a single results file that contains the average estimated generalization error of the classified, and the pooled standard deviation (for cv5 and cv10 errortypes).
This script requires that R be installed and in the search path. To install R visit: http://www.r-project.org/. Once R is installed, run R and excecute the command “install.packages(“randomForest”)”, then type q() to exit.
Usage: supervised_learning.py [options]
Outputs a ranking of features (e.g. OTUs) by importance, an estimation of the generalization error of the classifier, and the predicted class labels and posterior class probabilities according to the classifier.
Simple example of random forests classifier:
supervised_learning.py -i otu_table.biom -m Fasting_Map.txt -c BarcodeSequence -o ml
Running with 10-fold cross-validation for improved estimates of generalization error and feature importances:
supervised_learning.py -i otu_table.biom -m Fasting_Map.txt -c BarcodeSequence -o ml_cv10 -e cv10
Running with 1,000 trees for improved generalization error:
supervised_learning.py -i otu_table.biom -m Fasting_Map.txt -c BarcodeSequence -o ml_ntree1000 --ntree 1000
Run 10-fold cross validation on a directory of OTU tables rarefied at an even depth:
supervised_learning.py -i rarefied_tables/ -m Fasting_Map.txt -c Treatment -o sl_rarefied_tables_cv10 -e cv10
Run 10-fold cross validation on a directory of OTU tables rarefied at an even depth and collate the results into a single file:
supervised_learning.py -i rarefied_tables/ -m Fasting_Map.txt -c Treatment -o sl_rarefied_tables_cv10_sweep -e cv10 -w sl_cv10_sweep.txt