Skip to content
This repository has been archived by the owner on Apr 2, 2019. It is now read-only.

Home

Jump to bottom
Rhizosis edited this page Oct 2, 2013 · 21 revisions

Welcome to the Metrix wiki page

Wiki Links:

1. Introduction

With the release of the newer versions of software provided by Illumina, the creation of several high level summary files has been omitted. In the past Core Facilities employed these files in their pipeline purely as a summary or as information which was reported in their Laboratory Information Management Systems (LIMS).

The commonly used application supplied by Illumina called Sequence Analysis Viewer (SAV) is a handy tool to quickly give the user a visual insight in run efficiency and performance. Until now, low level inspection of data generated by Illumina's HiSeq 2000 and MiSeq machines was not possible because the data is stored in a efficient non-human readable binary format.

2. Metrix

Metrix was written to be able to access the binary files stored in the InterOp directory in each run folder. Written in JAVA metrix can be run cross platform and provides high level text summaries for core facility managers as well as low/mid level parsing methods to access distributions or 'lane, cycle and tile'-specific data layers for bioinformaticians.

To accomplish this Metrix supports multiple modes. These will be outlined in paragraphs 2.1 and 2.2.

2.1 Stand-alone mode

Many core facilities have the desire to obtain a high level summary after each run and can be run from the command line. This summary supplies the user with:

  • Percentage of Q20/Q30 reads
  • Q-Score distribution per lane
  • Phasing/Pre-phasing scores
  • Cluster Density / Cluster Density Passing Filter
  • Index Metrics overview
  • Error per lane
  • Current progress of the run (Current Cycle / Last Cycle)

2.2 Daemon mode

Using the daemon mode the core facility or the bioinformatics department can real time monitor and parse the storage directory of sequencing runs. If a run is started on a HiSeq or a MiSeq machine, Metrix will automatically parse the data and keep track of its progress. This way sequencing statistics can be easily integrated in data post processing pipelines, a LIMS or a home brew information tracking system. External applications can connect with Metrix using a server / client interface and interrogate all data available on the Metrix server using predefined methods. This way it has been made very easy to ask Metrix to return all sequencing runs with a specific state. These states include:

  • Currently active runs
  • Runs that stopped with errors
  • Paired-end runs that needs turning
  • Completed runs
  • Initializing runs (runs that have just been started)
  • All available runs in the database

In a later version of Metrix filtering methods can also be applied to limit the selection of data retrieved even more.

Clone this wiki locally