Introduction

LoFreq* (i.e. LoFreq version 2) is a fast and sensitive variant-caller for inferring SNVs and indels from next-generation sequencing data. It makes full use of base-call qualities and other sources of errors inherent in sequencing (e.g. mapping or base/indel alignment uncertainty), which are usually ignored by other methods or only used for filtering.

LoFreq* can run on almost any type of aligned sequencing data (e.g. Illumina, IonTorrent or Pacbio) since no machine- or sequencing-technology dependent thresholds are used. It automatically adapts to changes in coverage and sequencing quality and can therefore be applied to a variety of data-sets e.g. viral/quasispecies, bacterial, metagenomics or somatic data.

LoFreq* is very sensitive; most notably, it is able to predict variants below the average base-call quality (i.e. sequencing error rate). Each variant call is assigned a p-value which allows for rigorous false positive control. Even though it uses no approximations or heuristics, it is very efficient due to several runtime optimizations and also provides a (pseudo-)parallel implementation. LoFreq* is generic and fast enough to be applied to high-coverage data and large genomes. On a single processor it takes a minute to analyze Dengue genome sequencing data with nearly 4000X coverage, roughly one hour to call SNVs on a 600X coverage E.coli genome and also roughly an hour to run on a 100X coverage human exome dataset.

For more details on the original version of LoFreq see Wilm et al. (2012).