Use Of Unique Molecular Identifiers Research Articles

Unique Molecular Identifiers and Multiplexing Amplicons Maximize the Utility of Deep Sequencing To Critically Assess Population Diversity in RNA Viruses.

Next generation sequencing (NGS)/deep sequencing has become an important tool in the study of viruses. The use of unique molecular identifiers (UMI) can overcome the limitations of PCR errors and PCR-mediated recombination and reveal the true sampling depth of a viral population being sequenced in an NGS experiment. This approach of enhanced sequence data represents an ideal tool to study both high and low abundance drug resistance mutations and more generally to explore the genetic structure of viral populations. Central to the use of the UMI/Primer ID approach is the creation of a template consensus sequence (TCS) for each genome sequenced. Here we describe a series of experiments to validate several aspects of the Multiplexed Primer ID (MPID) sequencing approach using the MiSeq platform. We have evaluated how multiplexing of cDNA synthesis and amplicons affects the sampling depth of the viral population for each individual cDNA and amplicon to understand the relationship between broader genome coverage versus maximal sequencing depth. We have validated reproducibility of the MPID assay in the detection of minority mutations in viral genomes. We have also examined the determinants that allow sequencing reads of PCR recombinants to contaminate the final TCS data set and show how such contamination can be limited. Finally, we provide several examples where we have applied MPID to analyze features of minority variants and describe limits on their detection in viral populations of HIV-1 and SARS-CoV-2 to demonstrate the generalizable utility of this approach with any RNA virus.

Impact of high platelet turnover on the platelet transcriptome: Results from platelet RNA-sequencing in patients with sepsis.

Sepsis is associated with high platelet turnover and elevated levels of immature platelets. Changes in the platelet transcriptome and the specific impact of immature platelets on the platelet transcriptome remain unclear. Thus, this study sought to address whether and how elevated levels of immature platelets affect the platelet transcriptome in patients with sepsis. Blood samples were obtained from patients with sepsis requiring vasopressor therapy (n = 8) and from a control group of patients with stable coronary artery disease and otherwise similar demographic characteristics (n = 8). Immature platelet fraction (IPF) was determined on a Sysmex XE 2100 analyser and platelet function was tested by impedance aggregometry. RNA from leukocyte-depleted platelets was used for transcriptome analysis by Next Generation Sequencing integrating the use of unique molecular identifiers. IPF (median [interquartile range]) was significantly elevated in sepsis patients (6.4 [5.3-8.7] % vs. 3.6 [2.6-4.6] %, p = 0.005). Platelet function testing revealed no differences in adenosine diphosphate- or thrombin receptor activating peptide-induced platelet aggregation between control and sepsis patients. Putative circular RNA transcripts were decreased in platelets from septic patients. Leukocyte contamination defined by CD45 abundance levels in RNA-sequencing was absent in both groups. Principal component analysis of transcripts showed only partial overlap of clustering with IPF levels. RNA sequencing showed up-regulation of 524 and down-regulation of 118 genes in platelets from sepsis patients compared to controls. Upregulated genes were mostly related to catabolic processes and protein translation. Comparison to published platelet transcriptomes showed a large overlap of changes observed in sepsis and COVID-19 but not with reticulated platelets from healthy donors. Patients with sepsis appear to have a less degraded platelet transcriptome as indicated by increased levels of immature platelets and decreased levels of putative circular RNA transcripts. The present data suggests that increased protein translation is a characteristic mechanism of systemic inflammation.

Nuclease-Assisted, Multiplexed Minor-Allele Enrichment: Application in Liquid Biopsy of Cancer.

The use of next-generation sequencing (NGS) to profile genomic variation of individual cancer species is revolutionizing the practice of clinical oncology. In liquid biopsy of cancer, sequencing of circulating-free DNA (cfDNA) is gradually applied to all stages of cancer diagnosis and treatment, serving as complement or replacement of tissue biopsies. However, analysis of cfDNA obtained from blood draws still faces technical obstacles due in part to an excess of wild-type DNA originating from normal tissues and hematopoietic cells. The resulting low-level mutation abundance often falls below routine NGS detection sensitivity and limits reliable mutation identification that meets clinical sensitivity and specificity standards. Despite sample preparation advances that reduce sequencing error rates via use of unique molecular identifiers (molecular barcodes) and error-suppression algorithms, excessive amounts of sequencing are still required to detect mutations at allelic frequency levels below 1%. This requirement reduces throughput and increases cost.In this chapter, we describe a sensitive multiplex mutation detection method that enriches mutation-containing DNA during sample preparation, prior to sequencing, thereby increasing signal-to-noise ratios and providing low-level mutation detection without excessive sequencing depth. We couple targeted next-generation sequencing with wild-type DNA removal using Nuclease-assisted Minor-allele Enrichment using Probe Overlap, NaME-PrO, a recently developed method to eliminate wild-type sequences from multiple targets simultaneously. A step by step guide to library preparation and data analysis are provided as well as some precautions during the sample handling.

Highly Sensitive Detection Method of Retinoblastoma Genetic Predisposition and Biomarkers

Retinoblastoma is a malignant tumor of the infant retina. Nearly half of patients are predisposed to retinoblastoma by a germline RB1 pathogenic variant. Nonhereditary retinoblastoma is mainly caused by inactivation of both RB1 alleles at a somatic level. Several polymorphisms have been reported as biomarkers of retinoblastoma risk, aggressiveness, or invasion. The most informative genetic testing is obtained from tumor DNA. Historically, access to tumor DNA has been warranted by the frequent indication of enucleation, which has decreased because of advances in conservative approaches. Recent studies showed that tumor cell-free DNA can be analyzed in aqueous humor from retinoblastoma patients. This report describes a next-generation sequencing method relying on unique molecular identifiers for a highly sensitive detection of retinoblastoma genetic predisposition and biomarkers in a single analysis. It is the first use of unique molecular identifiers for retinoblastoma genetics. This gene panel enables the detection of RB1 point variants, large genome rearrangements, and loss of heterozygosity. It is adapted for genomic DNA extracted from blood or tumor DNA extracted from tumor fragment, aqueous humor, or plasma. The access to tumor cell-free DNA improves the diagnosis of genetic predisposition in case of conservative ocular therapy and provides access to biomarkers guiding the treatment strategy. The analysis of a gene panel is cost-effective and can be easily implemented in diagnostic laboratories.

Comparative analysis of targeted next-generation sequencing panels for the detection of gene mutations in chronic lymphocytic leukemia: an ERIC multi-center study.

Next-generation sequencing (NGS) has transitioned from research to clinical routine, yet the comparability of different technologies for mutation profiling remains an open question. We performed a European multicenter (n=6) evaluation of three amplicon-based NGS assays targeting 11 genes recurrently mutated in chronic lymphocytic leukemia. Each assay was assessed by two centers using 48 pre-characterized chronic lymphocytic leukemia samples; libraries were sequenced on the Illumina MiSeq instrument and bioinformatics analyses were centralized. Across all centers the median percentage of target reads ≥100x ranged from 94.2-99.8%. In order to rule out assay-specific technical variability, we first assessed variant calling at the individual assay level i.e., pairwise analysis of variants detected amongst partner centers. After filtering for variants present in the paired normal sample and removal of PCR/sequencing artefacts, the panels achieved 96.2% (Multiplicom), 97.7% (TruSeq) and 90% (HaloPlex) concordance at a variant allele frequency (VAF) >0.5%. Reproducibility was assessed by looking at the inter-laboratory variation in detecting mutations and 107 of 115 (93% concordance) mutations were detected by all six centers, while the remaining eight variants (7%) were undetected by a single center. Notably, 6 of 8 of these variants concerned minor subclonal mutations (VAF <5%). We sought to investigate low-frequency mutations further by using a high-sensitivity assay containing unique molecular identifiers, which confirmed the presence of several minor subclonal mutations. Thus, while amplicon-based approaches can be adopted for somatic mutation detection with VAF >5%, after rigorous validation, the use of unique molecular identifiers may be necessary to reach a higher sensitivity and ensure consistent and accurate detection of low-frequency variants.

UMI-VarCal: a new UMI-based variant caller that efficiently improves low-frequency variant detection in paired-end sequencing NGS libraries.

Next-generation sequencing has become the go-to standard method for the detection of single-nucleotide variants in tumor cells. The use of such technologies requires a PCR amplification step and a sequencing step, steps in which artifacts are introduced at very low frequencies. These artifacts are often confused with true low-frequency variants that can be found in tumor cells and cell-free DNA. The recent use of unique molecular identifiers (UMI) in targeted sequencing protocols has offered a trustworthy approach to filter out artefactual variants and accurately call low-frequency variants. However, the integration of UMI analysis in the variant calling process led to developing tools that are significantly slower and more memory consuming than raw-reads-based variant callers. We present UMI-VarCal, a UMI-based variant caller for targeted sequencing data with better sensitivity compared to other variant callers. Being developed with performance in mind, UMI-VarCal stands out from the crowd by being one of the few variant callers that do not rely on SAMtools to do their pileup. Instead, at its core runs an innovative homemade pileup algorithm specifically designed to treat the UMI tags in the reads. After the pileup, a Poisson statistical test is applied at every position to determine if the frequency of the variant is significantly higher than the background error noise. Finally, an analysis of UMI tags is performed, a strand bias and a homopolymer length filter are applied to achieve better accuracy. We illustrate the results obtained using UMI-VarCal through the sequencing of tumor samples and we show how UMI-VarCal is both faster and more sensitive than other publicly available solutions. The entire pipeline is available at https://gitlab.com/vincent-sater/umi-varcal-master under MIT license. Supplementary data are available at Bioinformatics online.

UMI-Gen: A UMI-based read simulator for variant calling evaluation in paired-end sequencing NGS libraries

MotivationWith Next Generation Sequencing becoming more affordable every year, NGS technologies asserted themselves as the fastest and most reliable way to detect Single Nucleotide Variants (SNV) and Copy Number Variations (CNV) in cancer patients. These technologies can be used to sequence DNA at very high depths thus allowing to detect abnormalities in tumor cells with very low frequencies. Multiple variant callers are publicly available and are usually efficient at calling out variants. However, when frequencies begin to drop under 1%, the specificity of these tools suffers greatly as true variants at very low frequencies can be easily confused with sequencing or PCR artifacts. The recent use of Unique Molecular Identifiers (UMI) in NGS experiments has offered a way to accurately separate true variants from artifacts. UMI-based variant callers are slowly replacing raw-read based variant callers as the standard method for an accurate detection of variants at very low frequencies. However, benchmarking done in the tools publication are usually realized on real biological data in which real variants are not known, making it difficult to assess their accuracy. ResultsWe present UMI-Gen, a UMI-based read simulator for targeted sequencing paired-end data. UMI-Gen generates reference reads covering the targeted regions at a user customizable depth. After that, using a number of control files, it estimates the background error rate at each position and then modifies the generated reads to mimic real biological data. Finally, it will insert real variants in the reads from a list provided by the user. AvailabilityThe entire pipeline is available at https://gitlab.com/vincent-sater/umigen under MIT license.

Mapping the Evolution of the Mutational Landscape of Acute Myeloid Leukemia from Diagnosis to Post-Transplantation Relapse and Its Interplay with Immune Evasion Mechanisms

Background Over the last decade, the advent of next-generation sequencing (NGS) has dramatically improved our understanding of the genomic landscape that characterize acute myeloid leukemia at time of diagnosis, leading to the identification of new prognostic subsets and therapeutic targets. Still, few studies have addressed how this landscape changes upon treatment, and in particular after allogeneic hematopoietic cell transplantation (allo-HCT). Moreover, the links between the mutational profile of AML and mechanisms of post-transplantation immune evasion remain to date largely unknown. Methods Taking advantage of the HaloPlex High Sensitivity (HS) technology, which allows a more precise definition of mutations and clonal architecture through the use of unique molecular identifiers, we designed and tested a NGS panel covering genes and miRNAs known to be involved in the pathogenesis of myeloid malignancies (n=112), in the DNA damage response (n=50) or in immune-related processes (n=30). By using an Illumina HiSeq2500n instrument,we sequenced a total of 200 AML samples collected from 131 patients, including 84 diagnoses, 45 relapses after chemotherapy (CT) and 71 relapses after allo-HCT, 31 of which characterized by genomic loss of HLA (Vago et al, N Engl J Med, 2009). Variant calling was performed using a pipeline based on the FreeBayes algorithm. Results Sequencing coverage was uniform across all target amplicons, with a 855x average depth-of-sequencing, and a mean of 122 unique barcodes for each region. Among the 84 diagnosis samples we identified 253 oncogenic mutations, with a median of 3 mutations per patient (range 0-8). Both the frequency and the prognostic value of mutations detected at diagnosis were in line with the largest published dataset on AML genomics (Papaemmanuil et al, N Engl J Med, 2016; R2= 0.88, p&amp;lt;0.0001). The overall burden of oncogenic mutations did not change significantly neither at relapses after CT nor after allo-HCT. Analyzing the profile of known oncogenic drivers of diagnoses and of the two relapse cohorts we documented high similarity between the three time-points, with significant differences limited to FLT3-ITD (present in 29% patients at diagnosis vs 49% and 48% at relapses after CT and allo-HSCT, p=0.02 and p=0.01 respectively) and WT1 mutations (5% vs 20% and 28%, p=0.006 and p&amp;lt;0.0001 respectively; Figure 1A). Also when comparing post-transplantation HLA loss relapses with their "classical" counterparts we could find no significant difference in the frequency or co-occurrence pattern of known AML driver mutations. Still, in HLA loss relapses we documented a significantly higher frequency of variants of unknown oncogenic potential befalling in genes implicated in the DNA damage response, including those belonging to the Fanconi Anemia pathway (35% of mutations in HLA loss vs 10% in classical relapses, p&amp;lt;0.0001; Figure 1B). This intriguing observation suggests that, in these cases, double strand breaks might be more frequently repaired through homologous recombination, and that this in turn might result in a higher likelihood of copy-neutral loss of heterozygosity events, the molecular mechanism at the basis of HLA loss. Finally, thanks to our panel, we were able to track possible routes of leukemic clonal evolution between diagnosis and post-transplantation relapse: interestingly, in several cases emergence of an HLA loss subclone out-competed clones carrying known oncogenic drivers, suggesting that after allo-HCT the immune privilege might outperform the proliferative advantage (Figure 1C). Discussion Taken together, our data highlight a complex interplay between the genomic landscape of AML and the environmental selective pressure represented by the graft-versus-leukemia effect. Whereas in fact some genomic alterations that carried little consequence before transplant become able in this setting to promote specific patterns of immune escape, others, that had driven AML biology before, turn out to be dispensable after allo-HCT. Understanding and tracking these evolutionary dynamics may provide new tools and targets to treat, or even prevent, AML post-transplantation relapses. Figure 1 Disclosures Sala: Gilead Company: Honoraria. Gentner:Genenta Science: Consultancy, Equity Ownership, Research Funding. Vago:GenDx: Research Funding; Moderna Therapeutics: Research Funding.

Systematic assessment of commercially available low-input miRNA library preparation kits

ABSTRACTHigh-throughput sequencing is increasingly favoured to assay the presence and abundance of microRNAs (miRNAs) in biological samples, even from low RNA amounts, and a number of commercial vendors now offer kits that allow miRNA sequencing from sub-nanogram (ng) inputs. Although biases introduced during library preparation have been documented, the relative performance of current reagent kits has not been investigated in detail. Here, six commercial kits capable of handling <100ng total RNA input were used for library preparation, performed by kit manufactures, on synthetic miRNAs of known quantities and human total RNA samples. We compared the performance of miRNA detection sensitivity, reliability, titration response and the ability to detect differentially expressed miRNAs. In addition, we assessed the use of unique molecular identifiers (UMI) sequence tags in one kit. We observed differences in detection sensitivity and ability to identify differentially expressed miRNAs between the kits, but none were able to detect the full repertoire of synthetic miRNAs. The reliability within the replicates of all kits was good, while larger differences were observed between the kits, although none could accurately quantify the relative levels of the majority of miRNAs. UMI tags, at least within the input ranges tested, offered little advantage to improve data utility. In conclusion, biases in miRNA abundance are heavily influenced by the kit used for library preparation, suggesting that comparisons of datasets prepared by different procedures should be made with caution. This article is intended to assist researchers select the most appropriate kit for their experimental conditions.

Abstract 5310: Novel, single-tube, multiplex rhPCR technology for detection of low-frequency variants in BRCA1 and BRCA2

Abstract Targeted next-generation sequencing (NGS) technologies serve as critical tools for cancer diagnostics. For example, individuals with pathological BRCA1 and BRCA2 (BRCA1/2) mutations have a dramatically increased risk for certain cancers, including breast and ovarian cancer. Those mutations are either germline or somatically acquired, and play an important role in cancer risk assessment, prevention of malignancy, and management of disease. A variety of methodologies are currently available to screen for BRCA1/2 mutations. Amplicon-based sequencing technologies are prevalent due to their easy workflow but have specific challenges due to inherent limitations of PCR, such as difficulties with amplification of overlapping amplicons for contiguous coverage in a single-tube or with detection of somatic low-frequency variants due to high polymerase error rates. We developed a robust, highly multiplex rhPCR workflow to screen for BRCA1/2 mutations across their entire coding regions in a single-tube. This method allows for detection of low-frequency variants through use of unique molecular identifiers (UMIs), which correct for PCR and sequencing errors and dramatically increase specificity. This method supports a wide DNA input range (1 to 50 ng) from cell-free, formalin-fixed paraffin-embedded (FFPE), or small biopsy samples. Our rhAmpSeq™ chemistry uses a two-enzyme system coupled with uniquely designed RNA-DNA hybrid primers, which enable the amplification of overlapping, multiplexed amplicons. Thus, we eliminate the need to split precious DNA samples into multiple tubes and prevent UMI hopping. Here, we demonstrate the performance of our novel, highly specific amplicon-based technology for detection of low-frequency variants of BRCA1/2 and compare our technology to current industry standards. Citation Format: Nicole Sponer, Yu Wang, Junzhou Wang, Shengyao Chen, Kevin Lai, Scott Rose, Patrick Lau. Novel, single-tube, multiplex rhPCR technology for detection of low-frequency variants in BRCA1 and BRCA2 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 5310.

Studying Cancer Heterogeneity by Single-Cell RNA Sequencing.

A major hurdle for the treatment of cancer is the incomplete understanding of its evolution through the course of its emergence, dispersal, and relapse. Genetic and epigenetic changes in combination with external cues and selective forces are the driving factors behind tumor heterogeneity. Understanding this variability within and across patients may partly explain the unpredictable outcomes of cancer treatments. Measuring the variation of gene expression levels within cells of the same tumor is a crucial part of this endeavor. Hence, the recently developed single-cell RNA-sequencing (scRNA-seq) technologies have become a valuable tool for cancer research. In practice, however, this is still challenging, especially for clinical samples. Here, we describe mcSCRB-seq (molecular crowding single-cell RNA barcoding and sequencing), a highly sensitive and powerful plate-based scRNA-seq method, which shows great capability to generate transcriptome data for cancer cells. mcSCRB-seq is not only characterized by high sensitivity due to molecular crowding and the use of unique molecular identifiers (UMIs) but also features an easy workflow and a low per-cell cost and does not require specialized equipment.

QsRNA-seq: a method for high-throughput profiling and quantifying small RNAs

The ability to profile and quantify small non-coding RNAs (sRNAs), specifically microRNAs (miRNAs), using high-throughput sequencing is challenging because of their small size. We developed QsRNA-seq, a method for preparation of sRNA libraries for high-throughput sequencing that overcomes this difficulty by enabling a gel-free separation of fragments shorter than 100 nt that differ only by 20 nt in length. The method allows the use of unique molecular identifiers for quantification and is more amenable to automation than gel-based methods. We show that QsRNA-seq gives very accurate, comprehensive, and reproducible results by looking at miRNAs in Caenorhabditis elegans embryos and larvae.

Abstract 3650: Use of unique molecular identifiers to gain insight about the true positive mutations in cfDNA of breast cancer patients for implementation of personalized treatment

Abstract Background: Blood analytes, as liquid biopsies, are discussed to be surrogate markers for therapy stratification, as serial sampling enabled by the minimal invasive nature of blood draw facilitates monitoring of clonal evolution. Mutational analysis of cell-free DNA (cfDNA) in plasma of breast cancer (BC) patients (pts) may predict the ideal therapy strategy. In this regard, PIK3CA mutations in BC pts' cfDNA were already shown to correlate with treatment response. However, in Next Generation Sequencing (NGS) workflows, used for cfDNA analysis, artefacts are often induced during library construction. Here we used unique molecular identifiers (UMIs) to verify true positive PIK3CA mutations in cfDNA of BC pts. Patients and methods: cfDNA was isolated from 54 plasma samples of 38 BC patients by affinity-based binding to magnetic beads (QIAamp MinElute ccfDNA Kit, QIAGEN, Germany). 30ng - 60ng cfDNA was used for library construction with the QIAseq Targeted DNA Panel for Illumina (QIAGEN, Germany) with integrated UMIs. Sequencing was executed on the NextSeq ® 500 platform (Illumina, US). Data analysis was performed by QIAseq Targeted Sequencing Data Analysis Portal and the Biomedical Genomics Workbench. As a reference, the PIK3CA mutational status of matched tumor tissue DNA (analyzed by Sanger sequencing) was consulted. Results: Library preparation was successful (yield &amp;gt;15ng) in 52/54 cases. Mean coverage was ~20.000x (mean UMI coverage ~2.500x) and &amp;gt;10.000x in 48/52 cases. The minimal allele frequency found for PIK3CA hotspot mutations (P539S, E545K, H1047R) by UMI analysis was 0.72%. In total, 133 mutations of the PIK3CA gene were identified as true positive mutations by UMI analysis in all 52 samples, which is a reduction of 69% (294/427) of all PIK3CA mutations incorrectly called by conventional data analysis. 59% of all different PIK3CA mutations called by UMI analysis appeared in over 5% of all pts. In the cohort of pts with PIK3CA hotspot mutant tumor tissue, 32% (6/19) showed the mentioned mutations also in matched cfDNA, whereas 16% of pts (3/19) without PIK3CA mutant tumor tissue were identified with true positive PIK3CA hotspot mutations in their plasma. Longitudinal analysis across two years during therapy revealed the increase in allele frequency (0%;11%;39%) of the PIK3CA H1047R mutation in one pt, whereas another pt showed a stable allele frequency of the PIK3CA P539S mutation (52%; 51%; 48%). Results will be expanded by consideration of mutations in BC hotspot genes despite PIK3CA and correlation to clinical parameter. Conclusions: Unique molecular identifiers enable the identification of true positive mutations in cfDNA and can thus, be used in clinical practice to determine molecular drivers of individual cancer progression and to employ personalized therapy. Citation Format: Corinna Keup, Karim Benyaa, Peter Hahn, Siegfried Hauch, Pawel Mach, Mitra Tewes, Hans-Christian Kolberg, Sabine Kasimir-Bauer. Use of unique molecular identifiers to gain insight about the true positive mutations in cfDNA of breast cancer patients for implementation of personalized treatment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3650.

Effect of method of deduplication on estimation of differential gene expression using RNA-seq.

BackgroundRNA-seq is a useful tool for analysis of gene expression. However, its robustness is greatly affected by a number of artifacts. One of them is the presence of duplicated reads.ResultsTo infer the influence of different methods of removal of duplicated reads on estimation of gene expression in cancer genomics, we analyzed paired samples of hepatocellular carcinoma (HCC) and non-tumor liver tissue. Four protocols of data analysis were applied to each sample: processing without deduplication, deduplication using a method implemented in SAMtools, and deduplication based on one or two molecular indices (MI). We also analyzed the influence of sequencing layout (single read or paired end) and read length. We found that deduplication without MI greatly affects estimated expression values; this effect is the most pronounced for highly expressed genes.ConclusionThe use of unique molecular identifiers greatly improves accuracy of RNA-seq analysis, especially for highly expressed genes. We developed a set of scripts that enable handling of MI and their incorporation into RNA-seq analysis pipelines. Deduplication without MI affects results of differential gene expression analysis, producing a high proportion of false negative results. The absence of duplicate read removal is biased towards false positives. In those cases where using MI is not possible, we recommend using paired-end sequencing layout.

Comparative Analysis of Single-Cell RNA Sequencing Methods

Single-cell RNA sequencing (scRNA-seq) offers new possibilities to address biological and medical questions. However, systematic comparisons of the performance of diverse scRNA-seq protocols are lacking. We generated data from 583 mouse embryonic stem cells to evaluate six prominent scRNA-seq methods: CEL-seq2, Drop-seq, MARS-seq, SCRB-seq, Smart-seq, and Smart-seq2. While Smart-seq2 detected the most genes per cell and across cells, CEL-seq2, Drop-seq, MARS-seq, and SCRB-seq quantified mRNA levels with less amplification noise due to the use of unique molecular identifiers (UMIs). Power simulations at different sequencing depths showed that Drop-seq is more cost-efficient for transcriptome quantification of large numbers of cells, while MARS-seq, SCRB-seq, and Smart-seq2 are more efficient when analyzing fewer cells. Our quantitative comparison offers the basis for an informed choice among six prominent scRNA-seq methods, and it provides a framework for benchmarking further improvements of scRNA-seq protocols.

The Use of Unique Molecular Identifiers (UMIs) Strongly Improves Sequencing Detection Limits Allowing Earlier Detection of Small TP53 Mutated Clones in Leukemias

Background:TP53 mutations arise in a broad set of hematologic diseases and are associated with poor prognosis and therapy failure. Even small TP53 mutated clones were demonstrated to be of clinical relevance and therefore their early detection is mandatory. Mutations can occur throughout the entire gene (mainly exons 4-10) and include base exchanges, deletions and insertions. Next generation sequencing (NGS) generally detects mutations, which are present in at least 3% of sequences. The detection of mutations at burdens below 3% is still hampered by polymerase and sequencing errors. Especially true mutations caused by single base exchanges (missense-, nonsense- and splice site mutations) are difficult to distinguish from non-specific background. To overcome this limitation, individual molecules can be tagged by unique molecular identifiers (UMIs). UMIs are small random sequences added to each individual molecule before amplification. Building a consensus sequence out of all amplified products derived from one original DNA template, allows reconstructing the sequence of the initial molecule and thereby eliminates amplification errors in silico.Aim: 1) To evaluate reduction of non-specific background in NGS assays after using UMIs to build consensus sequences. 2) To use consensus sequencing to identify early subclones by backtracking of known TP53 mutations in CLL patients.Methods: We added eight random nucleotides as UMI to all reverse primers (design adapted from Peng, et al. 2015) and performed an initial primer extension step. In a subsequent PCR, regions of interest including UMIs were amplified and adapters were added for MiSeq sequencing (Illumina, San Diego, CA). We performed 119 sequencing assays, and analyzed results with a 1% detection limit. Consensus sequences were built using SeqNext 4.3 (JSI Medical Systems, Kippenheim, Germany). A median number of 6 (range: 2-115) reads/consensus read was obtained. Samples contained 66 previously identified mutations (47 single base exchanges; 14 small deletions; 5 small insertions), of which 29 had a burden of 3% or lower. These had been confirmed to be specific by independently repeated sequencing analysis or known in patients at multiple time points during follow-up.Results: We evaluated background signal at each position before consensus read building. On average 97.5% of bases had low-level non-specific background. At each individual position, 0.01-0.87% of reads deviated from the reference sequence (Figure A). Building consensus reads by combining all sequences derived from one original DNA molecule reduced the number of bases with low-level non-specific signal to an average of 18% (Figure B).We performed 119 sequencing analysis with a 1% detection limit, aiming to identify 66 known mutations and no non-specific deviations from the reference sequence. With both approaches, 63/66 mutations were detected and correlation of mutation burdens was high (R2=0.99; P<0.001, calculated by linear regression). Two mutations (1% and 2%) only reached the detection cutoff without consensus read building and one mutation only after consensus read building. Importantly, without using UMIs for consensus reads, in sum 88 non-specific deviations from the reference sequence (likely artifacts) were detectable with the 1% cutoff, while only two likely artifacts remained after consensus sequences building.Thus, the reduction of sequencing artifacts should allow using a 1% cutoff for mutation detection in future routine settings. Therefore, we conducted a backtracking using 31 of the above identified TP53 mutations detected in 15 CLL patients. Samples covered a median timeframe of 26 (range: 2-68) months. Using the previous 3% cutoff for mutation burden, we identified 8/31 (26%) mutations 2-55 months earlier (median: 4 months). Even more, using the adjusted 1% cutoff with UMIs allowed detecting 18/31 (58%) TP53 mutations earlier (2-55 months; median: 18 months).Conclusion: Including UMIs and building consensus sequences, 1) reduced background signal in silico and 2) allowed improving NGS detection limits. This is crucial for the identification of low burden mutations in TP53 and other genes, where small subclones can rapidly expand and have been shown to require early treatment intervention. [Display omitted] DisclosuresBaer:MLL Munich Leukemia Laboratory: Employment. Nadarajah:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.