Observations vs. Taxa, filter_obs vs. filter_taxa, and Taxon_id vs. OTU_id.

[grabear]

Introduction

There is confusion amongst myself and others about what some of these words mean. This may be due to ignorance of the jargon used in the field. So I'm going to define them to the best of my abilitiy. I'm also going to attempt to make some suggestions to improve the readability of the package workflow and the syntax.

ropensci/taxa Class Diagram

Current Definitions

• Observations - User supplied data that has been formatted into a table. In the context of a Taxmap object, the observation data (e.g. OTUs, NCBI ids, etc) has also been mapped to the taxa/taxon/taxonomy objects which includes an edge list. This inclues anything in the obj$data list

  • filter_obs - This function filters the observation data found in the obj$data list. By default it attempts to leave the Taxa information untouched.

• Taxa - This is metacoder/taxa supplied information that has been interpolated from the annotations (e.g. Kingdom, Phylum, Class, etc). The taxa objects (taxa/taxon/taxonomy) are what create the network/tree used to relate the observations via taxon_id.

  • filter_taxa - This function filters the taxa data in the taxmap object. By default it attempts to filter the taxonomy data and the observation data.

• Taxon_id - The taxon_id is used to relate the observation data to the taxonomy object. The taxon_ids are what represent the nodes of the network/tree, and help link the annotation data and other observation data directly to these nodes.

• OTU_id - The otu_id is used outside of metacoder. While it can be used as a filtering mechanism there is currently no other use for this type of ID. (Correct me if I'm wrong.. I'm running out of time.)

[grabear]

Unfortunately, I've run out of time, but I will update this Issue when I get back.