ABEMUS: platform-specific and data-informed detection of somatic SNVs in cfDNA

Abstract

Motivation The use of liquid biopsies for cancer patients enables the non-invasive tracking of treatment response and tumor dynamics through single or serial blood drawn tests. Next-generation sequencing assays allow for the simultaneous interrogation of extended sets of somatic single-nucleotide variants (SNVs) in circulating cell-free DNA (cfDNA), a mixture of DNA molecules originating both from normal and tumor tissue cells. However, low circulating tumor DNA (ctDNA) fractions together with sequencing background noise and potential tumor heterogeneity challenge the ability to confidently call SNVs. Results We present a computational methodology, called Adaptive Base Error Model in Ultra-deep Sequencing data (ABEMUS), which combines platform-specific genetic knowledge and empirical signal to readily detect and quantify somatic SNVs in cfDNA. We tested the capability of our method to analyze data generated using different platforms with distinct sequencing error properties and we compared ABEMUS performances with other popular SNV callers on both synthetic and real cancer patients sequencing data. Results show that ABEMUS performs better in most of the tested conditions proving its reliability in calling low variant allele frequencies somatic SNVs in low ctDNA levels plasma samples. Availability and implementation ABEMUS is cross-platform and can be installed as R package. The source code is maintained on Github at http://github.com/cibiobcg/abemus, and it is also available at CRAN official R repository. Supplementary information Supplementary data are available at Bioinformatics online.

Precision cancer classification using liquid biopsy and advanced machine learning techniques

“…Researchers are actively exploring somatic variations in circulating cfDNA as a means of early cancer detection across a range of cancer types 13 , 15 , 16 , including gastric 17 , colorectal 18 , lung 19 , breast 20 , early-stage lung 21 , and late-stage human malignancies 22 . The ABEMUS 23 is geared towards identifying somatic single-nucleotide variants in cfDNA for both cancer detection and monitoring recurrent cancer growth using sequencing data from cancer patients 24 .…”

Circulating Tumor DNA in Head and Neck Squamous Cell Carcinoma

“…With the addition of unique molecular identifiers (UMI) cfDNA fragments can be tagged before amplification steps which enable in silico correction of these biases downstream. Computational algorithms such as iDES [ 123 ], PEC [ 125 ], TNER [ 126 ], ABEMUS [ 127 ] and SiNVICT [ 128 ] can correct for stereotypical PCR errors which become especially relevant at lower VAFs detection limits [ 123 ]. Although using different estimation models, most of these algorithms calculate and remove background mutation error counts based on healthy references.…”

Synggen: fast and data-driven generation of synthetic heterogeneous NGS cancer data

“…Reference models are built across a single or a collection of WES or TS sequencing control samples, profiled with the same platform, using a fast and efficient multi-threaded cumulative pileup strategy based on ( Valentini et al , 2019 ). Specifically, synggen generates: (i) a Read Depth Model ( RDM ), which measures the average intra-sample depth of coverage variability by calculating the probability, across all input files, of observing sequencing reads at any captured genomic region and position; only reads aligning at the lowest genomic coordinate are considered when paired-end data is used; (ii) a Quality Model ( QM ), which measures the distribution of base qualities across sequencing read positions; (iii) Position-Based Error ( PBE ), which measures for each captured genomic position [not representing a common single nucleotide polymorphism (SNP)] the probability of observing platform-specific systematic errors supported by high quality reads and bases ( Casiraghi et al , 2020 ). When paired-end data are used to generate the reference models, insert size statistics are also computed and embedded in the RDM model.…”