AMR Pipeline Workflow

Sign up for a CZ ID account and try the AMR workflow here! 

Overview

CZ ID's AMR pipeline implements the Resistance Gene Identifier (RGI) tool for AMR sequence detection. The RGI tool is used to compare quality controlled reads and assembled contigs against AMR references sequences from the Comprehensive Antibiotic Resistance Database (CARD). Below is an overview of the AMR pipeline workflow.

---

AMR Database

The CARD includes two database modules, namely the canonical CARD and a supplemental database known as CARD's Resistomes & Variants (nicknamed WildCARD). The canonical CARD includes curated and functionally characterized AMR genes from the literature, whereas WildCARD includes computationally predicted Resistomes and Variants and their Prevalence. The latter significantly expands CARD by providing more diverse reference sequences for RGI’s metagenomic read alignment algorithms compared to those available through the canonical CARD, which is strongly biased towards common clinical pathogens (see CARD 2023 research article for details). RGI, as implemented in the CZ ID AMR pipeline, uses WildCARD for predicting pathogen-of-origin for identified AMR genes.

All CZ ID AMR pipeline versions implement RGI version 6 (v6.0.0). However, the AMR pipeline version will determine which database version will be used to detect AMR genes. Samples uploaded to AMR pipeline v1.2.4 will be compared against canonical CARD v3.2.6 and WildCARD v4.0.0, whereas samples uploaded to prior versions will use canonical CARD v3.2.3 and WildCARD v3.1.0. Click here to follow CZ ID AMR pipeline changes and here to view canonical CARD and WildCARD (same as Prevalence, Resistomes, & Variants data) version details.

---

Get Started

See Upload or Select Data for AMR Analysis for details on how to run samples using the AMR pipeline. Note that there is an older AMR pipeline (known as the deprecated AMR pipeline) that implements a different workflow. See Current vs Deprecated AMR Pipelines to learn about differences between AMR pipelines. If you are interested in analyzing data with the older AMR workflow, you can still access results generated through the deprecated AMR pipeline (see Deprecated AMR Pipeline Results for details).