MOTIVATION: Calling copy number alterations (CNAs) from RNA sequencing (RNA-Seq) is challenging, because of the marked variability in coverage across genes and paucity of single nucleotide polymorphisms (SNPs). We have adapted SuperFreq to call absolute and allele sensitive CNAs from RNA-Seq. SuperFreq uses an error-propagation framework to combine and maximize information from read counts and B-allele frequencies. RESULTS: We used datasets from The Cancer Genome Atlas (TCGA) to assess the validity of CNA calls from RNA-Seq. When ploidy estimates were consistent, we found agreement with DNA SNP-arrays for over 98% of the genome for acute myeloid leukaemia (TCGA-AML, n = 116) and 87% for colorectal cancer (TCGA-CRC, n = 377). The sensitivity of CNA calling from RNA-Seq was dependent on gene density. Using RNA-Seq, SuperFreq detected 78% of CNA calls covering 100 or more genes with a precision of 94%. Recall dropped for focal events, but this also depended on signal intensity. For example, in the CRC cohort SuperFreq identified all cases (7/7) with high-level amplification of ERBB2, where the copy number was typically >20, but identified only 6% of cases (1/17) with moderate amplification of IGF2, which occurs over a smaller interval. SuperFreq offers an integrated platform for identification of CNAs and point mutations. As evidence of how SuperFreq can be applied, we used it to reproduce the established relationship between somatic mutation load and CNA profile in CRC using RNA-Seq alone. AVAILABILITY AND IMPLEMENTATION: SuperFreq is implemented in R and the code is available through GitHub: https://github.com/ChristofferFlensburg/SuperFreq/. Data and code to reproduce the figures are available at: https://gitlab.wehi.edu.au/flensburg.c/SuperFreq_RNA_paper. Data from TCGA (phs000178) was accessed from GDC following completion of a data access request through the database of Genotypes and Phenotypes (dbGaP). Data from the Leucegene consortium was downloaded from GEO (AML samples: GSE67040; normal CD34+ cells: GSE48846). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: RNA sequencing allows the study of both gene expression changes and transcribed mutations, providing a highly effective way to gain insight into cancer biology. When planning the sequencing of a large cohort of samples, library size is a fundamental factor affecting both the overall cost and the quality of the results. Here we specifically address how overall library size influences the detection of somatic mutations in RNA-seq data in two acute myeloid leukaemia datasets. RESULTS : We simulated shallower sequencing depths by downsampling 45 acute myeloid leukaemia samples (100 bp PE) that are part of the Leucegene project, which were originally sequenced at high depth. We compared the sensitivity of six methods of recovering validated mutations on the same samples. The methods compared are a combination of three popular callers (MuTect, VarScan, and VarDict) and two filtering strategies. We observed an incremental loss in sensitivity when simulating libraries of 80M, 50M, 40M, 30M and 20M fragments, with the largest loss detected with less than 30M fragments (below 90%, average loss of 7%). The sensitivity in recovering insertions and deletions varied markedly between callers, with VarDict showing the highest sensitivity (60%). Single nucleotide variant sensitivity is relatively consistent across methods, apart from MuTect, whose default filters need adjustment when using RNA-Seq. We also analysed 136 RNA-Seq samples from the TCGA-LAML cohort (50 bp PE) and assessed the change in sensitivity between the initial libraries (average 59M fragments) and after downsampling to 40M fragments. When considering single nucleotide variants in recurrently mutated myeloid genes we found a comparable performance, with a 6% average loss in sensitivity using 40M fragments. CONCLUSIONS: Between 30M and 40M 100 bp PE reads are needed to recover 90-95% of the initial variants on recurrently mutated myeloid genes. To extend this result to another cancer type, an exploration of the characteristics of its mutations and gene expression patterns is suggested.

Analysing multiple cancer samples from an individual patient can provide insight into the way the disease evolves.Monitoring the expansion and contraction of distinct clones helps to reveal the mutations that initiate the disease and those that drive progression.Existing approaches for clonal tracking from sequencing data typically require the user to combine multiple tools that are not purpose-built for this task.Furthermore, most methods require a matched normal (non-tumour) sample, which limits the scope of application.We developed SuperFreq, a cancer exome sequencing analysis pipeline that integrates identification of somatic single nucleotide variants (SNVs) and copy number alterations (CNAs) and clonal tracking for both.SuperFreq does not require a matched normal and instead relies on unrelated controls.When analysing multiple samples from a single patient, SuperFreq cross checks variant calls to improve clonal tracking, which helps to separate somatic from germline variants, and to resolve overlapping CNA calls.To demonstrate our software we analysed 304 cancer-normal exome samples across 33 cancer types in The Cancer Genome Atlas (TCGA) and evaluated the quality of the SNV and CNA calls.We simulated clonal evolution through in silico mixing of cancer and normal samples in known proportion.We found that SuperFreq identified 93% of clones with a cellular fraction of at least 50% and mutations were assigned to the correct clone with high recall and precision.In addition, SuperFreq maintained a similar level of performance for most aspects of the analysis when run without a matched normal.SuperFreq is highly versatile and can be applied in many different experimental settings for the analysis of exomes and other capture libraries.We demonstrate an application of SuperFreq to leukaemia patients with diagnosis and relapse samples. Author summaryCancer is a disease that continues to evolve.Understanding how it changes can provide key biological insights; for example, it can help to identify recurrent patterns associated with therapy resistance.However, tracking clonal evolution in a cancer from sequencing data is a major analytical challenge.We have developed SuperFreq, an analysis framework purpose built for the detection of intra-tumoural heterogeneity and clonal evolution.