This training will focus on analyzing QIAseq DNA panel datasets with QIAGEN CLC Genomics Workbench and the Biomedical Genomics Analysis plugin, including a live “FASTQ-to-VCF” demo of data import, data analysis and investigation. We will also show how to set up the analysis of a custom QIAseq DNA panel.
During this training, you’ll learn about:
• Import of FASTQ files
• Launching an analysis workflow
• Inspection of QC reports, genome browser view, detected variants and other workflow outputs
• Customization of template workflow parameters based on findings in the QC report
• Import of custom primers and target regions file
• How to set up an analysis of a custom panel
We’ll also have time for Q&A, so bring your questions to the training.
Recently, there have been many noteworthy papers citing QIAGEN CLC Genomics Workbench, a comprehensive, easy-to-use toolbox that ensures continuity in your NGS workflow. Here, we round up just a few of them to offer a sense of the diversity of the research for which QIAGEN CLC Genomics Workbench makes a difference. Below are some examples of how researchers from all over the world use this solution as a tool for metagenomic analysis to characterize dengue viruses and pathogens, create de novo assemblies or investigate ocular diseases.
Genomic characterization of SARS-CoV-2 identified in a reemerging COVID-19 outbreak in Beijing's Xinfadi market in 2020
First author: Yong Zhang
Should we be looking for new mutations in SARS-CoV-2 that make it more virulent? Researchers from the Chinese Center for Disease Control and Prevention perform genomic characterization of SARS-CoV-2 identified in a reemerging outbreak in China. Discover how they use QIAGEN CLC Genomics Workbench to help trace the source of the virus in this second outbreak in Beijing’s Xinfadi market. Read their full article here.
Genetic tracing of HCoV-19 for the re-emerging outbreak of COVID-19 in Beijing, China
First author: Jing Yang
Crucial coronavirus research from the Chinese Academy of Sciences looking into the re-emergence of the SARS-CoV-2 virus in China. Discover how they use the nanopore and MiSeq system together with QIAGEN CLC Genomics Workbench to trace the source of the virus in this second outbreak in Beijing. Get the full article here.
Systematic reconstruction of the complete two-component sensorial network in Staphylococcus aureus
First author: B. Rapun-Araiz
High-impact research by B. Rapun-Araiz and colleagues at Universidad Publica de Navarra in Spain who investigate the targets of two-component signal transduction systems (TCSs) in bacteria. See how they use QIAGEN CLC Genomics Workbench to map the complete TCS regulon in Staphylococcus aureus. Read the full paper here.
Remdesivir inhibits SARS-CoV-2 in human lung cells and chimeric SARS-CoV expressing the SARS-CoV-2 RNA polymerase in mice
First author: Andrea J. Pruijssers
Excellent research by A. Pruijssers and colleagues at Vanderbilt University who study how Remdesivir inhibits SARS-CoV-2 in human lung cells. See how they use QIAGEN CLC Main Workbench to help investigate the efficacy of Remdesivir against SARS-CoV-2 in vitro and in vivo. Read the full article here.
Lysosomal recycling of amino acids affects ER quality control
First author: Ryo Higuchi-Sanabria
Exciting research from the Howard Hughes Medical Institute, where researchers investigate the role of lysosomes in amino acid recycling. Learn how they use QIAGEN CLC Genomics Workbench to understand how reduced lysine and arginine can cause increased sensitivity to proteotoxic stress in the endoplasmic reticulum (ER). Read the full paper here.
Identifying SARS-CoV-2 related coronaviruses in Malayan pangolins
First author: Tommy Tsan-Yuk Lam
Coronavirus researchers from Hong Kong University use QIAGEN extraction kits and QIAGEN CLC Genomics Workbench to identify SARS-CoV-2 in Malayan pangolins. Their research helps reveal how pangolins may have facilitated the coronavirus transfer to humans, causing the COVID-19 disease. Read their Nature publication here.
Influenza A viruses are transmitted via the air from the nasal respiratory epithelium of ferrets
First author: Mathilde Richard
In honor of Global Hand Hygiene Day, remember to wash your hands! Check out this paper by researchers at Erasmus University Medical Center, who use QIAGEN CLC Genomics Workbench to investigate how influenza and other respiratory viruses are transmitted from nasal tracts using ferrets as a model. Read their full paper in Nature Communications.
Discovery of a subgenotype of human coronavirus NL63 associated with severe lower respiratory tract infection in China, 2018
First author: Yangun Wang
Learn about the critical research by Dr. Y. Wang and team from Guangzhou Medical University who studied a subgenotype of human coronavirus, NL63. They used QIAGEN CLC Genomics Workbench to investigate how this virus undergoes continuous mutation and has the potential to cause severe lower respiratory tract infection in humans. Read their research here.
Discovery of bat coronaviruses through surveillance and probe capture-based next-generation sequencing
First author: Bei Li
Dr. B. Li and colleagues from Wuhan Institute of Virology have been observing bats for potential coronavirus outbreaks after the SARS and MERS incidents. With the current pandemic, better surveillance practices are necessary to predict and mitigate the emergence of these viruses in humans. See how the team uses QIAGEN CLC Genomics Workbench and QIAGEN extraction kits in a capture-based NGS approach to overcome cost challenges. Discover their research here.
The splicing factor hnRNP M is a critical regulator of innate immune gene expression in macrophages
First author: Kelsi O. West
Great research from Texas A&M HSC where K. West and colleagues look at how pre-mRNA splicing decisions influence or are affected by macrophage activation. See how they use QIAGEN CLC Genomics and QIAGEN IPA to understand this link to the innate immune response in this Cell reports paper.
Microbiota dysbiosis and its pathophysiological significance in bowel obstruction
First author: Shrilakshmi Hedge
April is IBS awareness month. Check out this intriguing research by S. Hegde and colleagues from UTMB who look at how bowel obstruction may cause changes to the gut microbiota composition. See how the team utilizes a complete Sample to Insight approach using QIAGEN's extraction kits for bacterial DNA and RNA and QIAGEN CLC Microbial Genomics Module to identify bacterial species affected
Encapsulation boosts islet-cell signature in differentiating human induced pluripotent stem cells via integrin signaling
First author: Thomas Aga Legøy
Exciting research from the University of Bergen, where a team uses every part of the QIAGEN RNA-seq solution from Sample to Insight. See how QIAGEN CLC Genomics Workbench, QIAGEN IPA and other QIAGEN products help the team understand the development process of human-induced pluripotent stem cells into pancreatic islet cells. You can access the full Scientific Reports paper here.
Genetic aberrations in iPSCs are introduced by a transient G1/S cell cycle checkpoint deficiency
First author: Ryoko Araki
Crucial research for cell replacement therapy by Dr. R. Araki and colleagues from the National Institute of Radiological Sciences (NIRS) in Japan where they study how point mutations in reprogrammed pluripotent stem cells prevent their therapeutic application. Learn how the team uses QIAGEN CLC Genomics Workbench to understand how a cell cycle checkpoint deficiency causes a cancer-like state in these cells. Read the full Nature Sciences article here.
Applied shotgun metagenomics approach for the genetic characterization of dengue viruses
First author: Erley Lizarazo
Dengue virus (DENV) is the fastest pandemic-prone arthropod-borne virus, and is detected through virus serology, isolation of the virus or molecular identification. In this Science Direct paper, an international team of researchers optimized DENV detection using shotgun metagenomics. CLC Genomics Workbench was used to identify, genotype and characterize DENV in tested samples, including SNV calling. Importantly, researchers were able to identify multiple DENV serotypes in the same sample using CLC Genomics Workbench and have defined shotgun metagenomics as a suitable technique for detection and typing of DENV.
FDA-ARGOS is a database with public quality-controlled reference genomes
First author: Heike Sichtig
For correct microbial detection and identification by NGS, quality-controlled and tested databases are fundamental. In a Nature Communications paper, researchers from multiple US government labs and organizations, including NCBI, present the FDA-ARGOS quality-controlled reference genomes as a public database and demonstrate its utility in two example cases. In the first case, CLC Genomics Workbench was used to analyze sequencing reads. For metagenomic analysis, paired-end reads were trimmed and scored on the Phred scale, and trimmed reads were mapped to the Enterococcus avium assembly and Homo sapiens assembly using CLC genomics workbench. The researchers showed an accurate microbial identification of E. avium from metagenomic samples with the FDA-ARGOS reference genomes compared to non-curated GenBank genomes. For Ebola virus molecular inversion probes (MIPS), there was 100% concordance between the gold standard real-time PCR comparator and the in silico target sequence comparison, supporting the feasibility of this strategy for use in NGS-based assay evaluation studies.
A comparison of three different bioinformatics analyses of the 16S–23S rRNA encoding region for bacterial identification
First author: Nilay Peker
To optimize the development of antimicrobial therapy, rapid and reliable identification of pathogens from samples are required. Although Sanger sequencing of the 16S ribosomal RNA (rRNA) gene is used, species identification and discrimination are not always possible due to high sequence homology of the 16S rRNA gene among species. Recently, next-generation sequencing (NGS) of the 16S-23S rRNA encoding region has been proposed as a means for reliable identification of pathogens from samples. However, data analysis is time-consuming, and a database for the complete 16S-23S rRNA encoding regions is not available.
In this study, researchers from the University of Groningen in the Netherlands compared speed and accuracy of different data analysis approaches for 16S-23S rRNA NGS data: de novo assembly followed by BLAST, operational taxonomic unit (OTU) clustering or mapping, using an in-house developed 16S-23S rRNA encoding region database for identification of bacterial species. CLC Genomics Workbench was used for de novo assembly, mapping, and OTU clustering using the CLC Microbial Genomics Module. Furthermore, the researchers’ in-house developed 16S-23S rRNA database was uploaded to CLC Genomics Workbench. The researchers concluded that de novo assembly and BLAST appear to be the optimal approaches for data analysis, with the fastest turnaround time and highest sensitivity for sequencing the 16S-23S rRNA gene.
Role of oxidative stress in Retinitis pigmentosa: new involved pathways by an RNA-Seq analysis
First author: Luigi Donato
Retinitis pigmentosa (RP) is an inherited ocular disease characterized by progressive retinal disruption. One of the leading causes of RP is oxidative stress which arrests the metabolic support of photoreceptors. In this study, a group of researchers from Italy investigated the role of oxidative stress in RP onset and progression by whole transcriptome analysis of human retinal pigment epithelium cells, untreated or treated with 100 µg/ml oxLDL to induce oxidative stress. CLC Genomics Workbench was used for data analysis, including trimming of low-quality reads and quantification of gene expression. As a result, the researchers discovered 29 candidate genes associated with RP.
Request your no-obligation trial of QIAGEN CLC Genomics Workbench today!
Check out these recent articles citing Biomedical Genomics Workbench, a comprehensive, highly accurate NGS data analysis platform, providing researchers with a user-friendly, customizable human hereditary disease and cancer analysis solution for biomarker discovery and validation. Below are a few examples of how researchers from Pennsylvania to Japan are using Biomedical Genomics Workbench to accelerate their research.
Relaxin Reverses Inflammatory and Immune Signals in Aged Hearts
First author: Brian Martin
A team based out of the University of Pennsylvania studied the cardiovascular benefits of relaxin—a pregnancy hormone—on both young and old rats to determine its effects on the heart’s aging process. They extracted RNA and analyzed genomic changes, importing raw transcript data into Biomedical Genomics Workbench and mapping reads to the rat reference genome. The study, which ran in PLOS ONE, concluded that relaxin both alters gene transcription and suppresses inflammatory pathways and genes associated with heart failure and aging. This has therapeutic potential for cardiovascular and inflammation-related diseases, such as heart failure, diabetes and atrial fibrillation.
Comparison of Genetic Profiling of Primary Central Nervous System (CNS) Lymphoma Before and After Extra-CNS Relapse
First author: Kosuke Toyoda
In 2017, a team of Japanese scientists studied the mechanism of chemotherapy resistance in lymphomas of the CNS (central nervous system), which were previously identified as promising targets for immune checkpoint blockade therapy. They performed comprehensive genomic analysis in the hope of better understanding tumor oncogenic evolution and overcoming the immune privilege. The team compared the impact of extra-CNS relapse, using Biomedical Genomics Workbench to call variants. Their report, which ran in Blood Journal, suggested that the evolution of mutations enabled systemic disease progression with a breakthrough of immune privilege, characterized by immunological overpowering and the dysregulation of B-cell proliferation signaling.
Assessing the GeneRead SNP for Analysis of Low-Template and PCR-Inhibitory Samples
First author: Maja Sidstedt
When forensic DNA laboratories use massive parallel sequencing for human identification purposes, chances are good that the DNA samples are heterogeneous and of varying quality. SNP assays must therefore be able to handle impurities and low amounts of DNA. Using Biomedical Genomics Workbench to analyze sequencing data, a Swedish team evaluated the GeneRead Individual Identity SNP panel, which handled multiple extraction methods and withstood inhibitor solutions and was concluded to be satisfactory for casework-like samples. Read about the study, which ran in PLOS ONE in January this year.
To request your no-obligation trial of Biomedical Genomics Workbench, just click here.
The recent AMP Europe 2018 conference was a wonderful chance to catch up with old friends and establish new relationships—our team provided demos at the booth and we had a wonderful symposium. We also participated in a fun challenge, known as “Innovation Lab: Battle of the Bioinformatics Pipeline.” According to this story by Julia Karow in GenomeWeb, the aim of the exercise was “to provide commercial vendors of NGS analysis and interpretation pipelines with sequencing data from real patient samples, generated by a routine molecular diagnostics laboratory, and to see how similar or different their results would be.”
QIAGEN was one of three vendors who participated, using Biomedical Genomics Workbench data analysis platform and Qiagen Clinical Insight (QCI) Interpret software to identify mutations in tumor sequence data, down to a level of 5 percent. Participants were instructed to name and annotate the variants, state their allele frequencies and interpret them according to a five-tier classification system ranging from “benign” to “clinically significant.”
The session was organized and led by Winand Dinjens, head of molecular diagnostics in the Department of Pathology at Erasmus University Medical Center (Erasmus MC) Rotterdam, whose lab also analyzed the data, to establish a benchmark against which the other outcomes were compared. During the session, all three vendors presented their results and compared them to those of Erasmus MC. Though there was plenty of overlap amongst the three vendors’ results, none were identical. The session concluded with all participants agreeing that context (of a patient’s disease) is important in variant interpretation, and that laboratories must define their own thoughtful criteria to effectively frame a clinical report.
We are honored to have been included in the #AMPEurope2018 challenge, and that Biomedical Genomics Workbench and QCI were part of the process. We are also very proud of our team’s positive results—this is our third such challenge, 1) ECP 2017 and 2) AG MolPath, and we welcome the chance to compete again!
An estimated 300 million people worldwide live with some form of rare disease. In the US, a disease is considered rare if it affects fewer than 200,000 people, while in the European Union, rare disease affects fewer than 1 in 2,000 people. Advances are being made in ongoing research and in initiatives and communities that support patients, and QIAGEN is pleased to once again support Rare Disease Day 2018 — the theme of which is research.
Research conquers scientific frontiers and translates genomic insights into new medicines in the rare disease community. At QIAGEN, we offer a suite of solutions that contribute to these efforts, including Biomedical Genomics Workbench, Biomedical Genomics Server, and Ingenuity Variant Analysis. We are proud that our tools are helping scientists contribute to efforts to unravel these challenging diseases.
Our tools have recently been cited by researchers in their efforts to better understand rare disease. To learn more about rare inherited cardiac disorders—the primary cause of sudden cardiac death for those below the age of 35—Anders Krogh Broendberg and his team cited CLC Genomics Workbench as one of the bioinformatics tools used to call variants in their study. At the University of Paris, Lydie Da Costa used CLC Biomedical Workbench to analyze ribosomal protein genes inherent in Diamond-Blackfan anemia, a rare congenital bone marrow failure syndrome.
QIAGEN is proud to advocate for further research to help those with rare diseases, and we stand with scientists who strive to solve these complex genetic conundrums.
The dust has settled after a whirlwind annual meeting of the American Society of Human Genetics last month. The QIAGEN team would like to thank the many scientists who stopped by our booth to learn about how our bioinformatics tools can make a difference for their projects. We spent a lot of time exploring scientific posters at the conference and came away really impressed by how much great work is being done with tools such as IPA, QIAGEN Clinical Insight (QCI), Biomedical Genomics Workbench, and more.
Thanks to our video team, we have several short clips of researchers discussing some truly fascinating scientific results. Here’s a quick tour.
Jessa Hospital used IPA to understand response to therapy for patients with Crohn’s disease and inflammatory bowel disease.
An evaluation of QCI that helped his team cut variant interpretation times by 75 percent.
Using CLC Genomics Workbench and QCI to understand variants associated with intellectual disability in children.
The challenges of interpreting variants implicated in rare disease.
Using 37-gene QIAseq panels in a wide-ranging study of pancreatic cancer.
The utility of Biomedical Genomics Workbench for analyzing QIAseq panels.
Here at QIAGEN, we frequently fine-tune our solutions to better serve and support our customers in the international research and clinical communities, so they can continue to advance science and patient care. Changes range from minor tweaks — like bug fixes — to entirely new capabilities, like new templates or plugins. If you missed any of our recent updates about new features and capabilities of our line of bioinformatics solutions, here’s a brief roundup of some of the highlights you might want to know about.
This fall, we announced that our CLC Genomics Workbench 11 can be used as a genome browser to share, view and explore NGS analysis results, with no license required. This release also includes faster speeds, improved trimming and updated executables. We also released Biomedical Genomics Workbench 5, which debuted the QIAseq Targeted Panel Analysis plugin. This plugin enables accurate identification of genetic variants with ease, offers a user-friendly interface to simplify QIAseq data analysis, and introduces unique molecular indices and advanced algorithms to improve the accuracy of variant calling. The fall release of Ingenuity Variant Analysis included improvements to the Phenotype Driven Ranking feature by offering further sub-ranks for variants with identical scores. For QCI Interpret for Hereditary Cancer and QCI Interpret for Somatic Cancer, we introduced four new changes, including alignment of AMP/ASCO/CAP interpretation and reporting guidelines, increased flexibility, improved reporting templates and the ability for lab managers to set up groups. We also released updates that comprise the genome interpretation sector of our end-to-end sequencing solution: CLC Main Workbench, CLC Genomics Server 10, CLC Command Line Tools 5 and CLC Sequence Viewer 8.
Overall, we’re delighted to be ending 2017 with our solutions primed to take on even tougher bioinformatics challenges! If you’d like to learn more about one of these solutions or updates, please contact us here.
We’re just days away from heading to the annual meeting of the American Society of Human Genetics (ASHG) in Orlando, and we’re looking forward to checking in with customers and partners while at the show. Look for us at booth #745, where we’ll be exhibiting our sample-to-insight solutions. And while you’re perusing the posters, don’t miss the ones listed below — we think they’re particularly interesting.
(Bioinformatics and Computational Approaches)
Wednesday, Oct. 18 from 2.00 p.m. – 3.00 p.m.
Presenter: Bjarni Vilhjalmsson
To improve detection of low-frequency variants in cancer, QIAGEN scientists created a sample to insight solution using QIAseq targeted panels to incorporate unique molecular identifiers for NGS, followed by data analysis with the Biomedical Genomics Workbench software. Applying this innovative workflow to several data sets, the team significantly increased sequencing quality and achieved more accurate estimates of variant frequency.
(Epigenetics and Gene Regulation)
Wednesday, Oct. 18 from 3.00 p.m. – 4.00 p.m.
Presenter: Bethan Hussey
The first of two posters about epigenetic studies at Loughborough University in the United Kingdom, which examined the influence of exercise in altering DNA methylation and gene expression linked to inflammatory conditions. Using QIAGEN’s EpiTect LyseAll kits and Pyromark Q48 Autoprep assays, the scientists found significant changes in methylation of two genes post-exercise.
(Mendelian Phenotypes)
Wednesday, Oct. 18 from 3.00 – 4.00 p.m.
Presenter: Andreas Rump
Using NGS to investigate causes of delayed development in children, researchers at the University of Technology Dresden and University Clinic Leipzig in Germany relied on QIAGEN’s Biomedical Genomics Workbench software for variant calling, a critical step in interpreting genetic findings.
(Bioinformatics and Computational Approaches)
Thursday, Oct. 19 from 2.00 p.m. – 3.00 p.m.
Presenter: Tejaswi Koganti
Researchers at the Mayo Clinic in Rochester, Minn., tested CLC software for variant calling of small genetic variations known as indels. The study found QIAGEN’s CLC bioinformatics tools coupled with NGS delivered 95% accuracy in identifying insertions of fewer than 30 base pairs and deletions of fewer than 27 base pairs.
(Cancer Genetics)
Thursday, Oct. 19 from 3.00 p.m. – 4.00 p.m.
Presenter: Kambiz Karimi
Counsyl, a clinical laboratory in South San Francisco, Calif., compared QIAGEN Clinical Insight (QCI) software for interpretation of NGS results from 1,900 variants in hereditary cancer and other diseases to manual interpretation by PhD scientists and genetic counselors. The study found QCI’s coverage of variants was comprehensive and concordant with the lab’s own analysis. According to the poster, QCI facilitated significant time savings, freeing up lab staff time for difficult cases.
(Epigenetics and Gene Regulation)
Thursday, Oct. 19 from 3.00 p.m. – 4.00 p.m.
Presenter: David John Hunter
The second poster about epigenetic studies at Loughborough University in the United Kingdom, about the effects of exercise on DNA methylation and gene expression.
We hope to see you at the show!
Research scientist Dr. Margarete Odenthal leads the translational molecular pathology group at University Hospital Cologne, putting new technologies through their paces to determine which ones best fit the needs and workflows of the diagnostics lab. Recently, she has worked closely with QIAGEN to assess sample extraction, target enrichment, and library preparation kits as well as data analysis and interpretation tools in a variety of cancer studies.
In one NGS-based study, she and her team analyzed BRCA1 and BRCA2 mutations in samples from severe ovarian cancer using QIAGEN products for library preparation, amplicon sequencing, and data analysis. Odenthal began with macrodissected FFPE ovarian cancer samples, which carried known germline point mutations or large deletions in the BRCA genes, using the GeneRead DNAseq Targeted Panels for human BRCA1 and BRCA2 exons for target enrichment. After sequencing, results were analyzed with Biomedical Genomics Workbench to identify somatic pathogenic mutations. “We used the copy number variation tool in order to see big deletions in the BRCA1 and BRCA2 genes,” Odenthal says. “Normally you don’t see these deletions very easily, so people have found this quite interesting.” By using the analysis tool to detect large deletions, her team is able to quickly evaluate pathogenic mutations likely to damage the protein.
In other work, she has been focused on new approaches to make sense of tumor activity that cannot be explained by DNA mutations alone. “In these cases, the tumor driving force might be less about mutations and more about different expression and splicing patterns,” Odenthal says. “There are some transcripts which are alternatively spliced and have a more oncogenic version of the protein.” Having this information can be relevant for decisions about which therapy to use, so Odenthal has been using a cohort of prostate cancer samples as the foundation for studies of DNA and RNA together. “QIAGEN has good chemistry to see the DNA mutations and in parallel to look at splicing variants and expression,” she notes.
In this pipeline, she combines DNA and cDNA in a single sequencing run. “You do the mutation and expression analysis in one workflow and you have everything together. I think modern pathology has to have everything in one pipeline,” Odenthal says. She believes that running separate FISH, NGS, and DNA promoter methylation analysis workflows will not be sustainable as diagnostic labs continue to ramp up their capacity. “It is much more efficient to have one workflow and get all this information,” she adds.
To learn more about how Dr. Odenthal has used QIAGEN Bioinformatics tools, read the full case study here.
Photo credit: Uniklinik Köln
For few areas of genomics, do best practices evolve as quickly and continuously as for RNA-seq applications.
As a consequence of the rapid development within RNA-seq, researchers struggle to ensure that their analysis pipelines meet the latest standards. This typically means testing and integrating the best performers among a growing number of analysis solutions. And in the daily routine users often run a mix of different tools for the respective analysis step they perform best, from read mapping through isoform quantification to the detection of differential abundance.
RNA-seq analysis is a declared focus area for us. Users of CLC Genomics Workbench and Biomedical Genomics Workbench rely on us to constantly evaluate emerging bioinformatics approaches and integrate leading approaches into our solutions in a way that follows modern design control and quality assurance criteria.
We’re here sharing some of the recent improvements and underlying methods implemented into our RNA-seq solution.
Reads are simultaneously mapped to the genome and the transcriptome before being combined into a unified picture of expression. This brings two advantages:
Stranded library preparation reduces the uncertainty associated with assigning a read to an isoform, and can reveal antisense regulation of a gene. To account for imperfect efficiency of the protocol, reads are mapped to both strands, and those with a highest scoring alignment in the incorrect orientation are filtered away.
A typical RNA sequencing read could originate from several isoforms of a gene. An Expectation-Maximization algorithm, similar to that of RSEM, is used to determine the actual expressed isoform. The algorithm generalizes the human intuition that if some reads can only originate from isoform A or B, and others can only originate from isoform B or C, then it is most likely that B is expressed.
Results on the benchmark of Teng et al. (2016) showing the consistency of isoform quantification (left) and the accuracy of detected fold changes (right) for several open source methods and for QIAGEN. In both plots, the x-axis shows the expression level, and the performance of all methods improves for isoforms with higher expression. Perfect quantification would be the line y=0 for the plot on the left, and y=1 for the plot on the right. Flux Capacitor, used by the GTEx consortium, performs worst on these benchmarks.
The QIAGEN quantification approach overlaps with the best performing open source tool, RSEM. Data for the open source methods were kindly provided by the authors of the benchmark.
Samples can be associated with editable metadata at any stage of analysis. The information in the metadata is flowing automatically into statistics and visualizations to provide insight into patterns of expression and the presence of confounding factors.
Intuitive analysis of differential expression between samples or sample groups. Results are depicted in tables, 2-D heat maps, or Venn diagrams. The results are linked and selecting a set of differentially expressed features reveals the respective information in other tables or plots. Differential expression of genes or transcripts can be compared among samples in the genome browser view (track list).
The genome browser view makes it easy to visualize genes or transcripts that are differentially expressed between samples. Color coding reveals differences in gene expression. Fold changes in expression are log-transformed and converted into color space.
Differential expression is detected based on metadata associated with samples. The statistics are based on the fit of a Generalized Linear Model with a negative binomial distribution, similar to EdgeR or DESeq2. The model supports paired designs, and can control for batch effects.
Differential expression can also be detected in exploratory studies where no replicates are available. In this case the algorithm shares data between isoforms with similar expression to estimate technical and biological variability.
We have carried out a comparative study to ensure that results generated with our solutions are in line with results generated with leading alternative methods.
Comparison to the DESeq2 benchmark of Love et al. (2014). This plot is equivalent to two panels of figure 6 of the paper by Love et al. but with QIAGEN added, and edgeR modified to run with quasi-likelihood testing. Data are simulated such that 20% of isoforms have a three-fold change in an experiment with 3 or 5 replicates (left and right plots respectively). The experiment is repeated six times. The sensitivity is the fraction of differentially expressed isoforms that are detected. The false discovery rate (FDR) is the fraction of isoforms that the method incorrectly calls as being differentially expressed. A perfect method would have highest possible sensitivity while lying on the black line (which is the target FDR, here set to 0.1). On this data DESeq and edgeR lie to the left of the target error rate, meaning they are being too conservative. DESeq2 and QIAGEN performs best on this benchmark with DESeq2 performing favourably with lower numbers of replicates. Differences between methods become smaller as the size of the fold change or number of replicates increase.
Teng, M., Love, M. I., Davis, C. A., Djebali, S., Dobin, A., et al. (2016). A benchmark for RNA-seq quantification pipelines. Genome Biology, 17(1), 74. doi:10.1186/s13059-016-0940-1
Love, M. I., Huber, W., & Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550. doi:10.1186/s13059-014-0550-8
