Skip to content

SMuRF 3.0

Jump to bottom
EvaPang2022 edited this page Feb 6, 2024 · 1 revision

Introducing SMuRF 3.0

Machine learning-based R package for consensus somatic mutation prediction

STRELKA2 - MUTECT2 - FREEBAYES - VARDICT - VARSCAN

Introduction

SMuRF 3.0.0 incorporates the output of 5 somatic variant callers from the bcbio-nextgen 1.2.9 pipeline. Users without access to Strelka2 outputs may download SMuRF v1.6.4 here.

Updates

  • Compatible with bcbio-nextgen 1.2.9 pipeline

Fig.1

Performance of SMuRF 3.0. Precision-recall profiles for individual somatic mutation callers and SMuRF evaluated on SNV and INDELs using 20% withheld test data. SMuRF models were trained on 80% matched tumour-normal WGS data from a chronic lymphocytic leukemia (CLL) patient and a medulloblastoma (MB) patient (Alioto et al., Nat. Comms 2015). To generate variations in sequencing coverage and tumour purity, noise was injected into the training and testing data; 1: normal samples down-sampled to 30x coverage, 2: lowered purity tumour samples spiked in with either 12x, 30x and 42x normal reads respectively (80%, 50% and 30% purity).

DREAM Benchmarks

  • Benchmark using independent data from the DREAM Somatic Mutation Challenge (Ewing et al., Nat. Methods 2015) shows that SMuRF 3.0 performs better than MuTect2, Strelka2, VarScan, VarDict and FreeBayes.

Fig.3

Fig.1

Fig.2

Legend: SMuRF-3.0 (red), SMuRF-v2.0.12 (black)

Clone this wiki locally