HUMID: reference free FastQ deduplication

---

HUMID is a tool to quickly and easily remove duplicate reads from FastQ files, with or without UMIs.

Installation

You can install HUMID from conda

conda install -c bioconda humid

If you want to, you can also install HUMID from source.

Usage

Both the input and output of HUMID are plain FastQ files, no alignment required! This means that you can use HUMID to remove duplicates as a pre-processing step before starting your analysis. If your project was sequenced without UMIs, or if the UMIs are present in the headers of the FastQ reads (as is done by BCL Convert), you can use the following command:

humid forward.fastq.gz reverse.fastq.gz

If the UMIs are located in a separate FastQ file use

humid forward.fastq.gz reverse.fastq.gz umi.fast.gz

For other use cases, we recommend that you use fastp to move the UMIs to the header of the forward FastQ file before deduplicating them with HUMID.

Please see the usage section of the documentation for more details.

Contents:

• Introduction
  • UMI deduplication
• Installation
  • From source
• Command Line Interface
• Usage
  • Word length
  • UMIs in separate file
  • UMIs in the header with underscore
  • UMIs in the header with colon
  • Other UMI formats
  • Deduplication without UMI
• Contributors