Amount Of Input DNA Research Articles

Abstract 1233: Reducing sequence artifact in clinical sequencing of treatment-naïve NSCLC patient using molecular barcoding system

Abstract Background: Recently the clinical significance of compound mutations in cancer relapse and/or acquired resistance has been reported. Amplicon-based deep sequence enables analysis of tiny amount of input DNA, however, a major problem of these high-throughput DNA sequencing is the high rate (~1%) of errors causing potential sequence artifact. Molecular barcoding system, which has been developed to reduce sequencing artifacts as well as to improve mutation detection accuracy, was applied to analyze the frequency of compound EGFR mutations in early stage non-small-cell lung cancer (NSCLC) patients who undergo surgical resection without any presurgical treatment is less described. Materials and Methods: From 590 consecutive patients, 64 adenocarcinoma cases who underwent surgical resection for primary lung cancer were analyzed by using amplicon-based targeted sequencing method incorporating molecular barcodes in order to detect genetic alterations of 47 genes including EGFR. Results: Out of 64 samples, EGFR common mutation profiles of 63 (98.4%) by molecular-barcode sequencing corresponded to those by clinical test. Uncommon EGFR mutations were detected in 7 cases (10.9%). Among the three types of EGFR major mutation, G719X (60%, 3/5) showed a significantly higher incidence of EGFR double mutations than L858R (9.5%, 4/42) or Ex19del (0%, 0/17) (p = 0.0052). Co-mutations of other genes were observed in 20 EGFR-mutated cases. TP53 mutations were frequently detected in younger age (p = 0.0066) and pStage II-III cases (p = 0.042). Conclusion: Amplicon sequencing incorporating molecular barcoding system is a feasible approach to characterize predictive or prognostic mutations in early stage treatment naïve NSCLC patients, revealing those case who harbor EGFR G719X mutation have a significantly higher incidence of EGFR double mutations, likely to have worse prognosis. Citation Format: Kei Namba, Shuta Tomida, Yuta Takahashi, Eisuke Kurihara, Yusuke Ogoshi, Takahiro Yosioka, Hidejiro Torigoe, Hiroki Sato, Kazuhiko Shien, Hiromasa Yamamoto, Junichi Soh, Shinichi Toyooka. Reducing sequence artifact in clinical sequencing of treatment-naïve NSCLC patient using molecular barcoding system [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 1233.

Systematic evaluation of the early access applied biosystems precision ID Globalfiler mixture ID and Globalfiler NGS STR panels for the ion S5 system

Systematic evaluation of the early access applied biosystems precision ID Globalfiler mixture ID and Globalfiler NGS STR panels for the ion S5 system

Shallow whole genome sequencing for robust copy number profiling of formalin-fixed paraffin-embedded breast cancers

Pathology archives with linked clinical data are an invaluable resource for translational research, with the limitation that most cancer samples are formalin-fixed paraffin-embedded (FFPE) tissues. Therefore, FFPE tissues are an important resource for genomic profiling studies but are under-utilised due to the low amount and quality of extracted nucleic acids. We profiled the copy number landscape of 356 breast cancer patients using DNA extracted FFPE tissues by shallow whole genome sequencing. We generated a total of 491 sequencing libraries from 2 kits and obtained data from 98.4% of libraries with 86.4% being of good quality. We generated libraries from as low as 3.8 ng of input DNA and found that the success was independent of input DNA amount and quality, processing site and age of the fixed tissues. Since copy number alterations (CNA) play a major role in breast cancer, it is imperative that we are able to use FFPE archives and we have shown in this study that sWGS is a robust method to do such profiling.

Evaluating droplet digital PCR for the quantification of human genomic DNA: converting copies per nanoliter to nanograms nuclear DNA per microliter.

The highly multiplexed polymerase chain reaction (PCR) assays used for forensic human identification perform best when used with an accurately determined quantity of input DNA. To help ensure the reliable performance of these assays, we are developing a certified reference material (CRM) for calibrating human genomic DNA working standards. To enable sharing information over time and place, CRMs must provide accurate and stable values that are metrologically traceable to a common reference. We have shown that droplet digital PCR (ddPCR) limiting dilution end-point measurements of the concentration of DNA copies per volume of sample can be traceably linked to the International System of Units (SI). Unlike values assigned using conventional relationships between ultraviolet absorbance and DNA mass concentration, entity-based ddPCR measurements are expected to be stable over time. However, the forensic community expects DNA quantity to be stated in terms of mass concentration rather than entity concentration. The transformation can be accomplished given SI-traceable values and uncertainties for the number of nucleotide bases per human haploid genome equivalent (HHGE) and the average molar mass of a nucleotide monomer in the DNA polymer. This report presents the considerations required to establish the metrological traceability of ddPCR-based mass concentration estimates of human nuclear DNA. Graphical abstract The roots of metrological traceability for human nuclear DNA mass concentration results. Values for the factors in blue must be established experimentally. Values for the factors in red have been established from authoritative source materials. HHGE stands for "haploid human genome equivalent"; there are two HHGE per diploid human genome.

Estimating mutation load at 5% LOD from FFPE samples using a targeted next-generation sequencing assay.

28 Background: High tumor mutation load is a biomarker which positively correlates with response to immune checkpoint inhibitors. Current methods to estimate tumor mutation load often require large amounts of DNA. Herein, we demonstrate the ability of a targeted panel to estimate mutation load from FFPE samples using low input amount of DNA. Methods: We developed a single-sample analysis workflow to estimate mutation load (somatic mutation count per Megabase (Mb)) from FFPE and fresh frozen tumor research samples. The assay utilizes a PCR-based target enrichment panel that covers ~1.7 Mb. Our workflow requires only 10 ng of input DNA, and enables a 2.5-day turn-around time from sample to the final report. The workflow enables < 60 minutes of hands-on time for automated library preparation and templating on a batch of 8 samples. Sequencing is performed using a high throughput semiconductor sequencing platform to achieve sufficient depth (~500x coverage) and accuracy. Our analysis pipeline calls variants with optimized parameters on the tumor sample only, with no matched normal sample required, and applies filters to remove germ-line. Results: An in-silico analysis using 21,000 exomes from COSMIC demonstrated the panel can achieve high sensitivity (>=93%) and specificity (>99%) necessary to stratify high and low mutation burden samples. Matched tumor-normal analyses on 14 lung and colorectal samples showed that our single tumor analysis workflow detects mutation load with strong correlation (r=0.8699) with tumor-normal analysis. Through dilution experiments on engineered control we learned that the workflow provides accurate estimates of mutation load, predicting ± 7% of expected mutation load on >=20% tumor content. To assess reproducibility, we compared mutation load in library replicates for a cohort of 8 samples (FFPE and fresh frozen tumors, engineered control, and NIST cell-lines) and observed high correlation (r=0.9906) among replicates. Conclusions: We developed a simple workflow on the Ion Torrent sequencing platform to estimate mutation load from FFPE and fresh frozen tumor research samples. This solution will advance research in immuno-oncology.

Assessment of Capture and Amplicon-Based Approaches for the Development of a Targeted Next-Generation Sequencing Pipeline to Personalize Lymphoma Management

Targeted next-generation sequencing panels are increasingly used to assess the value of gene mutations for clinical diagnostic purposes. For assay development, amplicon-based methods have been preferentially used on the basis of short preparation time and small DNA input amounts. However, capture sequencing has emerged as an alternative approach because of high testing accuracy. We compared capture hybridization and amplicon sequencing approaches using fresh-frozen and formalin-fixed, paraffin-embedded tumor samples from eight lymphoma patients. Next, we developed a targeted sequencing pipeline using a 32-gene panel for accurate detection of actionable mutations in formalin-fixed, paraffin-embedded tumor samples of the most common lymphocytic malignancies: chronic lymphocytic leukemia, diffuse large B-cell lymphoma, and follicular lymphoma. We show that hybrid capture is superior to amplicon sequencing by providing deep more uniform coverage and yielding higher sensitivity for variant calling. Sanger sequencing of 588 variants identified specificity limits of thresholds for mutation calling, and orthogonal validation on 66 cases indicated 93% concordance with whole-genome sequencing. The developed pipeline and assay identified at least one actionable mutation in 91% of tumors from 219 lymphoma patients and revealed subtype-specific mutation patterns and frequencies consistent with the literature. This pipeline is an accurate and sensitive method for identifying actionable gene mutations in routinely acquired biopsy materials, suggesting further assessment of capture-based assays in the context of personalized lymphoma management.

A robust targeted sequencing approach for low input and variable quality DNA from clinical samples

Next-generation deep sequencing of gene panels is being adopted as a diagnostic test to identify actionable mutations in cancer patient samples. However, clinical samples, such as formalin-fixed, paraffin-embedded specimens, frequently provide low quantities of degraded, poor quality DNA. To overcome these issues, many sequencing assays rely on extensive PCR amplification leading to an accumulation of bias and artifacts. Thus, there is a need for a targeted sequencing assay that performs well with DNA of low quality and quantity without relying on extensive PCR amplification. We evaluate the performance of a targeted sequencing assay based on Oligonucleotide Selective Sequencing, which permits the enrichment of genes and regions of interest and the identification of sequence variants from low amounts of damaged DNA. This assay utilizes a repair process adapted to clinical FFPE samples, followed by adaptor ligation to single stranded DNA and a primer-based capture technique. Our approach generates sequence libraries of high fidelity with reduced reliance on extensive PCR amplification—this facilitates the accurate assessment of copy number alterations in addition to delivering accurate single nucleotide variant and insertion/deletion detection. We apply this method to capture and sequence the exons of a panel of 130 cancer-related genes, from which we obtain high read coverage uniformity across the targeted regions at starting input DNA amounts as low as 10 ng per sample. We demonstrate the performance using a series of reference DNA samples, and by identifying sequence variants in DNA from matched clinical samples originating from different tissue types.

Generation of Whole Genome Bisulfite Sequencing Libraries for Comprehensive DNA Methylome Analysis.

Whole genome bisulfite sequencing (WGBS) enables the detection of DNA methylation at single base-pair resolution. The treatment of DNA with sodium bisulfite allows the discrimination of methylated and unmethylated cytosines, but the power of this technology can be limited by the input amounts of DNA and the length of DNA fragments due to DNA damage caused by the desulfonation process. Here, we describe a WGBS library preparation protocol that minimizes the loss and damage of DNA, generating high quality libraries amplified with fewer PCR cycles, and hence data with fewer PCR duplicates, from lower amounts of input material. Briefly, genomic DNA is sheared, end-repaired, 3'-adenylated, and ligated to adaptors with fewer cleanup steps in between, minimizing DNA loss. The adapter-ligated DNA is then treated with sodium bisulfite and amplified with few PCR cycles to reach the yield needed for sequencing.

Quantification of DNA in Neonatal Dried Blood Spots by Adenine Tandem Mass Spectrometry.

Newborn screening programs have expanded to include molecular-based assays as first-tier tests and the success of these assays depends on the quality and yield of DNA extracted from neonatal dried blood spots (DBS). To meet high throughput and rapid turnaround time requirements, newborn screening laboratories adopted rapid DNA extraction methods that produce crude extracts. Quantification of DNA in neonatal DBS is not routinely performed due to technical challenges; however, this may enhance the performance of assays that are sensitive to amounts of input DNA. In this study, we developed a novel high throughput method to quantify total DNA in DBS. It is based on specific acid-catalyzed depurination of DNA followed by mass spectrometric quantification of adenine. The amount of adenine was used to calculate DNA quantity per 3.2 mm DBS. Reference intervals were established using archived, neonatal DBS (n = 501) and a median of 130.6 ng of DNA per DBS was obtained, which is in agreement with literature values. The intra- and interday variations were <15%. The limits of detection and quantification were 12.5 and 37.8 nmol/L adenine, respectively. We demonstrated that DNA from neonatal DBS can be successfully quantified in high throughput settings using instruments currently deployed in NBS laboratories.

Comprehensive Whole DNA Methylome Analysis by Integrating MeDIP-seq and MRE-seq.

Understanding the role of DNA methylation often requires accurate assessment and comparison of these modifications in a genome-wide fashion. Sequencing-based DNA methylation profiling provides an unprecedented opportunity to map and compare complete DNA CpG methylomes. These include whole genome bisulfite sequencing (WGBS), Reduced-Representation Bisulfite-Sequencing (RRBS), and enrichment-based methods such as MeDIP-seq, MBD-seq, and MRE-seq. An investigator needs a method that is flexible with the quantity of input DNA, provides the appropriate balance among genomic CpG coverage, resolution, quantitative accuracy, and cost, and comes with robust bioinformatics software for analyzing the data. In this chapter, we describe four protocols that combine state-of-the-art experimental strategies with state-of-the-art computational algorithms to achieve this goal. We first introduce two experimental methods that are complementary to each other. MeDIP-seq, or methylation-dependent immunoprecipitation followed by sequencing, uses an anti-methylcytidine antibody to enrich for methylated DNA fragments, and uses massively parallel sequencing to reveal identity of enriched DNA. MRE-seq, or methylation-sensitive restriction enzyme digestion followed by sequencing, relies on a collection of restriction enzymes that recognize CpG containing sequence motifs, but only cut when the CpG is unmethylated. Digested DNA fragments enrich for unmethylated CpGs at their ends, and these CpGs are revealed by massively parallel sequencing. The two computational methods both implement advanced statistical algorithms that integrate MeDIP-seq and MRE-seq data. M&M is a statistical framework to detect differentially methylated regions between two samples. methylCRF is a machine learning framework that predicts CpG methylation levels at single CpG resolution, thus raising the resolution and coverage of MeDIP-seq and MRE-seq to a comparable level of WGBS, but only incurring a cost of less than 5% of WGBS. Together these methods form an effective, robust, and affordable platform for the investigation of genome-wide DNA methylation.

Whole-Genome Bisulfite Sequencing Using the Ovation® Ultralow Methyl-Seq Protocol.

The analysis of genome-wide epigenomic alterations including DNA methylation has become a subject of intensive research for many complex diseases. Whole-genome bisulfite sequencing (WGBS) using next-generation sequencing technologies can be considered the gold standard for a comprehensive and quantitative analysis of cytosine methylation throughout the genome. Several approaches including tagmentation- and post bisulfite adaptor tagging (PBAT)-based WGBS have been devised. Here, we provide a detailed protocol based on a commercial kit for the preparation of libraries for WGBS from limited amounts of input DNA (50-100ng) using the classical approach of WGBS by ligation of methylated adaptors to the fragmented DNA prior to bisulfite conversion. The converted library is then amplified with an optimal number of PCR cycles to ensure high sequence diversity and low duplicate rates. Spike-in of unmethylated DNA allows for the precise estimation of bisulfite conversion rates. We also provide a step-by-step description of the data analysis using publicly available bioinformatic tools. The described protocol has been successfully applied to different human samples as well as DNA extracted from plant tissues and yields robust and reproducible results.

Forensic DNA methylation profiling from minimal traces: How low can we go?

Analysis of human DNA methylation (DNAm) can provide additional investigative leads in crime cases, e.g. the type of tissue or body fluid, the chronological age of an individual, and differentiation between identical twins. In contrast to the genetic profile, the DNAm level is not the same in every cell. At the single cell level, DNAm represents a binary event at a defined CpG site (methylated versus non-methylated). The DNAm level from a DNA extract however represents the average level of methylation of the CpG of interest of all molecules in the forensic sample. The variance of DNAm levels between replicates is often attributed to technological issues, i.e. degradation of DNA due to bisulfite treatment, preferential amplification of DNA, and amplification failure. On the other hand, we show that stochastic variations can lead to gross fluctuation in the analysis of methylation levels in samples with low DNA levels. This stochasticity in DNAm results is relevant since low DNA amounts (1pg – 1ng) is rather the norm than the exception when analyzing forensic DNA samples.This study describes a conceptual analysis of DNAm profiling and its dependence on the amount of input DNA. We took a close look at the variation of DNAm analysis due to DNA input and its consequences for different DNAm-based forensic applications.As can be expected, the 95%-confidence interval of measured DNAm becomes narrower with increasing amounts of DNA. We compared this aspect for two different DNAm-based forensic applications: body fluid identification and chronological age determination. Our study shows that DNA amount should be well considered when using DNAm for forensic applications.

A method for measuring the distribution of the shortest telomeres in cells and tissues

Improved methods to measure the shortest (not just average) telomere lengths (TLs) are needed. We developed Telomere Shortest Length Assay (TeSLA), a technique that detects telomeres from all chromosome ends from <1 kb to 18 kb using small amounts of input DNA. TeSLA improves the specificity and efficiency of TL measurements that is facilitated by user friendly image-processing software to automatically detect and annotate band sizes, calculate average TL, as well as the percent of the shortest telomeres. Compared with other TL measurement methods, TeSLA provides more information about the shortest telomeres. The length of telomeres was measured longitudinally in peripheral blood mononuclear cells during human aging, in tissues during colon cancer progression, in telomere-related diseases such as idiopathic pulmonary fibrosis, as well as in mice and other organisms. The results indicate that TeSLA is a robust method that provides a better understanding of the shortest length of telomeres.

Fusobacterium nucleatum in gastroenterological cancer: Evaluation of measurement methods using quantitative polymerase chain reaction and a literature review.

The human microbiome Fusobacterium nucleatum, which primarily inhabits the oral cavity, causes periodontal disease and has also been implicated in the development of colorectal cancer. However, whether F. nucleatum is present in other gastroenterological cancer tissues remains to be elucidated. The present study evaluated whether quantitative polymerase chain reaction (qPCR) assays were able to detect F. nucleatum DNA and measure the quantity of F. nucleatum DNA in esophageal, gastric, pancreatic and liver cancer tissues. The accuracy of the qPCR assay was determined from a calibration curve using DNA extracted from cells from the oral cavity. Formalin-fixed paraffin-embedded (FFPE) tumor tissues from 20 patients with gastroenterological [esophageal (squamous cell carcinoma), gastric, colorectal, pancreatic and liver] cancer and 20 matched normal tissues were evaluated for F. nucleatum DNA content. The cycle threshold values in the qPCR assay for F. nucleatum and solute carrier organic anion transporter family member 2A1 (reference sample) decreased linearly with the quantity of input DNA (r2>0.99). The F. nucleatum detection rate in esophageal, gastric and colorectal cancer tissues were 20% (4/20), 10% (2/20) and 45% (9/20), respectively. F. nucleatum was not detected in liver and pancreatic cancer tissues. The qPCR results from the frozen and FFPE tissues were consistent. Notably, F. nucleatum was detected at a higher level in superficial areas compared with the invasive areas. F. nucleatum in esophageal, gastric and colorectal cancer tissues was evaluated by qPCR using FFPE tissues. F. nucleatum may be involved in the development of esophageal, gastric and colorectal cancer.

Comparison of standard capillary electrophoresis based genotyping method and ForenSeq DNA Signature Prep kit (Illumina) on a set of challenging samples

DNA extracted from the challenging samples was analyzed with a standard capillary electrophoresis (CE) based genotyping method and with the ForenSeq™ DNA Signature Prep kit on the Illumina MiSeq FGx™ Forensic Genomics system. The results showed that more data can be obtained by MPS compared with CE with the same amount of input DNA and that use of MPS is influenced by the degree of damage of the template and amount of DNA available.

Initial assessment of the Precision ID Globalfiler Mixture ID panel on the Ion Torrent S5XL DNA sequencer and Converge v2.0 software

The U.S. National Institute of Standards and Technology (NIST) participated in an assessment of the Precision ID Globalfiler early access Mixture ID Panel (v1.0), which consists of primers for amplification of over 100 forensically relevant loci in the human genome. Markers included in the panel can be divided into four classes: Short Tandem Repeats (STR) (29 autosomal STRs and 1 Y STR), single nucleotide polymorphisms (SNPs) (42 autosomal and two Y chromosome), insertion/deletion polymorphisms (Indels) (Amelogenin and Y Indel rs2032678), and microhaplotype blocks (MH) (36 clusters of two to four SNPs). Several challenging sample types were sequenced and analyzed, including: artificially degraded DNA, multiple-contributor mixtures, mixtures of related individuals, and picogram input amounts of DNA. Performance characteristics, in the form of mixture deconvolution outcomes and likelihood ratio calculations, for the Mixture ID Panel STRs relative to state-of-the-art capillary electrophoresis methods are presented here.

A MBD-seq protocol for large-scale methylome-wide studies with (very) low amounts of DNA

ABSTRACTWe recently showed that, after optimization, our methyl-CpG binding domain sequencing (MBD-seq) application approximates the methylome-wide coverage obtained with whole-genome bisulfite sequencing (WGB-seq), but at a cost that enables adequately powered large-scale association studies. A prior drawback of MBD-seq is the relatively large amount of genomic DNA (ideally >1 µg) required to obtain high-quality data. Biomaterials are typically expensive to collect, provide a finite amount of DNA, and may simply not yield sufficient starting material. The ability to use low amounts of DNA will increase the breadth and number of studies that can be conducted. Therefore, we further optimized the enrichment step. With this low starting material protocol, MBD-seq performed equally well, or better, than the protocol requiring ample starting material (>1 µg). Using only 15 ng of DNA as input, there is minimal loss in data quality, achieving 93% of the coverage of WGB-seq (with standard amounts of input DNA) at similar false/positive rates. Furthermore, across a large number of genomic features, the MBD-seq methylation profiles closely tracked those observed for WGB-seq with even slightly larger effect sizes. This suggests that MBD-seq provides similar information about the methylome and classifies methylation status somewhat more accurately. Performance decreases with <15 ng DNA as starting material but, even with as little as 5 ng, MBD-seq still achieves 90% of the coverage of WGB-seq with comparable genome-wide methylation profiles. Thus, the proposed protocol is an attractive option for adequately powered and cost-effective methylome-wide investigations using (very) low amounts of DNA.

Evaluation of the Precision ID Identity Panel for the Ion Torrent™ PGM™ sequencer

In cases where only a partial or incomplete STR profile is obtained from a sample, information contained in single nucleotide polymorphisms (SNPs) can prove informative for human identification. Thermo Fisher Scientific, which developed the high throughput Ion Torrent™ PGM™ sequencer, released the Precision ID Identity Panel, a multiplex SNP panel for human identity. We evaluated the reproducibility and sensitivity of this multiplex, which contains primers for the amplification of 90 autosomal SNPs and 34 Y-clade SNPs. The manufacturer’s protocol was tested using five commercially available pure native DNAs and six forensic type samples at a range of DNA input amounts (0.2–1.0ng; n, 90). In addition to analyzing the data using the manufacturer’s software, HID SNP Genotyper (v4.3.1), we also used CLC Genomics Workbench (Qiagen). Although library yields and templating of ion sphere particles (ISPs) were low, downstream sequencing was still successful. Across all samples, only 1.5% of all possible quality control (QC) flags were raised by both the plugin QC filter and CLC; 85% of those flags were raised as the SNP had a major allele frequency outside the thresholds specified by the manufacturer. For the remaining SNPs, coverage of >1500 X and >780 X was obtained for autosomal and Y-clade SNPs respectively, and 100% congruence among genotype calls from both analysis programs was observed. Our results demonstrate that it is possible to obtain reliable and reproducible genotypes using the Precision ID Identity Panel, when using low quantities (≥0.2ng) of either pure native DNA or forensic type DNA samples.

Comparative analysis of 12 different kits for bisulfite conversion of circulating cell-free DNA

ABSTRACTBlood circulating cell-free DNA (cfDNA) is becoming popular in the search of promising predictive and prognostic biomarkers. Among these biomarkers, cfDNA methylation markers have especially gained considerable attention. A significant challenge in the utilization of cfDNA methylation markers is the limited amount of cfDNA available for analyses; reportedly, bisulfite conversion (BSC) reduce cfDNA amounts even further. Nevertheless, few efforts have focused on ensuring high cfDNA conversion efficiency and recovery after BSC. To compare cfDNA recovery of different BSC methods, we compared 12 different commercially available BSC kits. We tested whether DNA recovery was affected by the molecular weight and/or quantity of input DNA. We also tested BSC efficiency for each kit. We found that recovery varied for DNA fragments of different lengths: certain kits recovered short fragments better than others, and only 3 kits recovered DNA fragments of <100 bp well. In contrast, DNA input amount did not seem to affect DNA recovery: for quantities spanning between 820 and ∼25,000 genome equivalents per BSC, a linear relation was found between input and recovery amount. Overall, mean recovery ranged between 9 and 32%, with BSC efficiency of 97–99.9%. When plasma cfDNA was used as input for BSC, recovery varied from 22% for the poorest and 66% for the best performing kits, while conversion efficiency ranged from 96 to 100% among different kits. In conclusion, clear performance differences exist between commercially available BSC kits, both in terms of DNA recovery and conversion efficiency. The choice of BSC kit can substantially impact the amount of converted cfDNA available for downstream analysis, which is critical in a cfDNA methylation marker setting.

Denoising genome-wide histone ChIP-seq with convolutional neural networks.

MotivationChromatin immune-precipitation sequencing (ChIP-seq) experiments are commonly used to obtain genome-wide profiles of histone modifications associated with different types of functional genomic elements. However, the quality of histone ChIP-seq data is affected by many experimental parameters such as the amount of input DNA, antibody specificity, ChIP enrichment and sequencing depth. Making accurate inferences from chromatin profiling experiments that involve diverse experimental parameters is challenging.ResultsWe introduce a convolutional denoising algorithm, Coda, that uses convolutional neural networks to learn a mapping from suboptimal to high-quality histone ChIP-seq data. This overcomes various sources of noise and variability, substantially enhancing and recovering signal when applied to low-quality chromatin profiling datasets across individuals, cell types and species. Our method has the potential to improve data quality at reduced costs. More broadly, this approach—using a high-dimensional discriminative model to encode a generative noise process—is generally applicable to other biological domains where it is easy to generate noisy data but difficult to analytically characterize the noise or underlying data distribution.Availability and implementation https://github.com/kundajelab/coda.