Portal de Periódicos da CAPES

Rapid Next-Generation Sequencing–Based Diagnostics of Bacteremia in Septic Patients

2020; Elsevier BV; Volume: 22; Issue: 3 Linguagem: Inglês

10.1016/j.jmoldx.2019.12.006

1943-7811

Christian Grumaz, Anne Hoffmann, Yevhen Vainshtein, Maria Kopp, Silke Grumaz, Philip Stevens, Sebastian Decker, Markus A. Weigand, Stefan Hofer, Thorsten Brenner, Kai Sohn,

Genomics and Phylogenetic Studies

The increasing incidence of bloodstream infections including sepsis is a major challenge in intensive care units worldwide. However, current diagnostics for pathogen identification mainly depend on culture- and molecular-based approaches, which are not satisfactory regarding specificity, sensitivity, and time to diagnosis. Herein, we established a complete diagnostic workflow for real-time high-throughput sequencing of cell-free DNA from plasma based on nanopore sequencing for the detection of the causative agents, which was applied to the analyses of eight samples from four septic patients and three healthy controls, and subsequently validated against standard next-generation sequencing results. By optimization of library preparation protocols for short fragments with low input amounts, a 3.5-fold increase in sequencing throughput could be achieved. With tailored bioinformatics workflows, all eight septic patient samples were found to be positive for relevant pathogens. When considering time to diagnosis, pathogens were identified within minutes after start of sequencing. Moreover, an extrapolation of real-time sequencing performance on a cohort of 239 septic patient samples revealed that more than 90% of pathogen hits would have also been detected using the optimized MinION workflow. Reliable identification of pathogens based on circulating cell-free DNA sequencing using optimized workflows and real-time nanopore-based sequencing can be accomplished within 5 to 6 hours following blood draw. Therefore, this approach might provide therapy-relevant results in a clinically critical timeframe. The increasing incidence of bloodstream infections including sepsis is a major challenge in intensive care units worldwide. However, current diagnostics for pathogen identification mainly depend on culture- and molecular-based approaches, which are not satisfactory regarding specificity, sensitivity, and time to diagnosis. Herein, we established a complete diagnostic workflow for real-time high-throughput sequencing of cell-free DNA from plasma based on nanopore sequencing for the detection of the causative agents, which was applied to the analyses of eight samples from four septic patients and three healthy controls, and subsequently validated against standard next-generation sequencing results. By optimization of library preparation protocols for short fragments with low input amounts, a 3.5-fold increase in sequencing throughput could be achieved. With tailored bioinformatics workflows, all eight septic patient samples were found to be positive for relevant pathogens. When considering time to diagnosis, pathogens were identified within minutes after start of sequencing. Moreover, an extrapolation of real-time sequencing performance on a cohort of 239 septic patient samples revealed that more than 90% of pathogen hits would have also been detected using the optimized MinION workflow. Reliable identification of pathogens based on circulating cell-free DNA sequencing using optimized workflows and real-time nanopore-based sequencing can be accomplished within 5 to 6 hours following blood draw. Therefore, this approach might provide therapy-relevant results in a clinically critical timeframe. Bloodstream infections, in particular sepsis, represent one of the main causes of death, with a mortality rate of up to 30% to 50% in intensive care units worldwide—and this trend is rising.1Fleischmann C. Scherag A. Adhikari N.K.J. Hartog C.S. Tsaganos T. Schlattmann P. Angus D.C. Reinhart K. International Forum of Acute Care TrialistsAssessment of global incidence and mortality of hospital-treated sepsis. Current estimates and limitations.Am J Respir Crit Care Med. 2016; 193: 259-272Crossref PubMed Scopus (1292) Google Scholar, 2Stevenson E.K. Rubenstein A.R. Radin G.T. Wiener R.S. Walkey A.J. Two decades of mortality trends among patients with severe sepsis: a comparative meta-analysis*.Crit Care Med. 2014; 42: 625-631Crossref PubMed Scopus (427) Google Scholar, 3Vincent J.L. Marshall J.C. Namendys-Silva S.A. Francois B. Martin-Loeches I. Lipman J. Reinhart K. Antonelli M. Pickkers P. Njimi H. Jimenez E. Sakr Y. ICON InvestigatorsAssessment of the worldwide burden of critical illness: the intensive care over nations (ICON) audit.Lancet Respir Med. 2014; 2: 380-386Abstract Full Text Full Text PDF PubMed Scopus (513) Google Scholar, 4Alberto L. Marshall A.P. Walker R. Aitken L.M. Screening for sepsis in general hospitalized patients: a systematic review.J Hosp Infect. 2017; 96: 305-315Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar To provide an appropriate treatment for sepsis patients, the major challenge is not only the timely diagnosis of sepsis, but also the identification of the sepsis-causing agents.4Alberto L. Marshall A.P. Walker R. Aitken L.M. Screening for sepsis in general hospitalized patients: a systematic review.J Hosp Infect. 2017; 96: 305-315Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar A rapid detection of the causative pathogens enables early targeted antimicrobial therapy, which significantly increases survival rate, prevents subsequent complications, and reduces drug-related side effects as well as medical expenses.4Alberto L. Marshall A.P. Walker R. Aitken L.M. Screening for sepsis in general hospitalized patients: a systematic review.J Hosp Infect. 2017; 96: 305-315Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar, 5David V.L. Ercisli M.F. Rogobete A.F. Boia E.S. Horhat R. Nitu R. Diaconu M.M. Pirtea L. Ciuca I. Horhat D. Horhat F.G. Licker M. Popovici S.E. Tanasescu S. Tataru C. Early prediction of sepsis incidence in critically ill patients using specific genetic polymorphisms.Biochem Genet. 2017; 55: 193-203Crossref PubMed Scopus (17) Google Scholar, 6Soong J. Soni N. Sepsis: recognition and treatment.Clin Med (Lond). 2012; 12: 276-280Crossref PubMed Scopus (50) Google Scholar, 7Dellinger R.P. Levy M.M. Rhodes A. Annane D. Gerlach H. Opal S.M. Sevransky J.E. Sprung C.L. Douglas I.S. Jaeschke R. Osborn T.M. Nunnally M.E. Townsend S.R. Reinhart K. Kleinpell R.M. Angus D.C. Deutschman C.S. Machado F.R. Rubenfeld G.D. Webb S.A. Beale R.J. Vincent J.L. Moreno R. Surviving Sepsis Campaign Guidelines Committee including the Pediatric SubgroupSurviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2012.Crit Care Med. 2013; 41: 580-637Crossref PubMed Scopus (4160) Google Scholar, 8Kumar A. Roberts D. Wood K.E. Light B. Parrillo J.E. Sharma S. Suppes R. Feinstein D. Zanotti S. Taiberg L. Gurka D. Kumar A. Cheang M. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock.Crit Care Med. 2006; 34: 1589-1596Crossref PubMed Scopus (3670) Google Scholar Despite extensive medical research in the field of sepsis, there are to date no fully satisfactory diagnostic approaches for a rapid, reliable, and sensitive identification of pathogens in the bloodstream.9Chun K. Syndergaard C. Damas C. Trubey R. Mukindaraj A. Qian S. Jin X. Breslow S. Niemz A. Sepsis pathogen identification.J Lab Autom. 2015; 20: 539-561Crossref PubMed Scopus (25) Google Scholar, 10Liesenfeld O. Lehman L. Hunfield K.-P. Kost G. Molecular diagnosis of sepsis: new aspects and recent developments.Eur J Microbiol Immunol (Bp). 2014; 4: 1-25Crossref PubMed Google Scholar, 11Lebovitz E.E. Burbelo P.D. Commercial multiplex technologies for the microbiological diagnosis of sepsis.Mol Diagn Ther. 2013; 17: 221-231Crossref PubMed Scopus (20) Google Scholar, 12Burillo A. Bouza E. Use of rapid diagnostic techniques in ICU patients with infections.BMC Infect Dis. 2014; 14: 593Crossref PubMed Scopus (25) Google Scholar Therefore, time-consuming, error- and contamination-prone blood cultures (BC) are still considered the standard of care for sepsis diagnostics, frequently leading to an inappropriate and delayed targeted therapy.11Lebovitz E.E. Burbelo P.D. Commercial multiplex technologies for the microbiological diagnosis of sepsis.Mol Diagn Ther. 2013; 17: 221-231Crossref PubMed Scopus (20) Google Scholar,13Minasyan H. Sepsis and septic shock: pathogenesis and treatment perspectives.J Crit Care. 2017; 40: 229-242Crossref PubMed Scopus (56) Google Scholar Recently, a clinically validated workflow for identification and quantification of sepsis-causing pathogens was established based on circulating cell-free DNA (cfDNA) from plasma by high-throughput sequencing.14Grumaz S. Stevens P. Grumaz C. Decker S.O. Weigand M.A. Hofer S. Brenner T. von Haeseler A. Sohn K. Next-generation sequencing diagnostics of bacteremia in septic patients.Genome Med. 2016; 8: 73Crossref PubMed Scopus (124) Google Scholar, 15Decker S.O. Sigl A. Grumaz C. Stevens P. Vainshtein Y. Zimmermann S. Weigand M.A. Hofer S. Sohn K. Brenner T. Immune-response patterns and next generation sequencing diagnostics for the detection of mycoses in patients with septic shock—results of a combined clinical and experimental investigation.Int J Mol Sci. 2017; 18: 1796Crossref Scopus (22) Google Scholar, 16Grumaz S. Grumaz C. Vainshtein Y. Stevens P. Glanz K. Decker S.O. Hofer S. Weigand M.A. Brenner T. Sohn K. Enhanced performance of next-generation sequencing diagnostics compared with standard of care microbiological diagnostics in patients suffering from septic shock.Crit Care Med. 2019; 47: e394-e402Crossref PubMed Scopus (17) Google Scholar Using this workflow, microbial species and genes relevant for antibiotic resistances could be identified within 24 to 30 hours after blood draw. Furthermore, by applying a newly established relevance score, the so-called sepsis indicating quantifier (SIQ) score, clinically relevant pathogens could be discriminated from nonrelevant contaminants based on statistical calculations. Using this novel approach, the drawbacks and limitations with regard to specificity and sensitivity of conventional BC and molecular-based methods (eg, PCR-based approaches10Liesenfeld O. Lehman L. Hunfield K.-P. Kost G. Molecular diagnosis of sepsis: new aspects and recent developments.Eur J Microbiol Immunol (Bp). 2014; 4: 1-25Crossref PubMed Google Scholar) could be largely overcome. Although such an unbiased workflow seems to be more sensitive and specific, there is still an urgent need for speeding up time to diagnosis. In the last decade, sequencing technologies developed rapidly, allowing accurate and economical high-throughput sequencing.17Goodwin S. McPherson J.D. McCombie W.R. Coming of age: ten years of next-generation sequencing technologies.Nat Rev Genet. 2016; 17: 333-351Crossref PubMed Scopus (1665) Google Scholar One of the most recent sequencing platforms is represented by the MinION (Oxford Nanopore Technologies, Oxford, UK), a handheld sequencer that is based on single-molecule sequencing using nanopores.18Clarke J. Wu H.-C. Jayasinghe L. Patel A. Reid S. Bayley H. Continuous base identification for single-molecule nanopore DNA sequencing.Nat Nanotechnol. 2009; 4: 265Crossref PubMed Scopus (1163) Google Scholar,19Mikheyev A.S. Tin M.M. A first look at the Oxford Nanopore MinION sequencer.Mol Ecol Resour. 2014; 14: 1097-1102Crossref PubMed Scopus (235) Google Scholar Despite its limited throughput and moderate sequencing accuracy to date, several outstanding features such as its portability, read out of ultra-long reads (>100 kb), and the immediate processing of generated reads in real time make this technology highly promising for many different applications.20Leggett R.M. Clark M.D. A world of opportunities with nanopore sequencing.J Exp Bot. 2017; 68: 5419-5429Crossref PubMed Scopus (77) Google Scholar Of interest, the potential of this sequencing platform with respect to clinical microbiology has already been shown for several studies ranging from outbreak surveillances to infection diagnostics from various clinical samples (urine, blood, implants, feces, sputum) and antimicrobial resistance profiling.21Greninger A.L. Naccache S.N. Federman S. Yu G. Mbala P. Bres V. Stryke D. Bouquet J. Somasekar S. Linnen J.M. Rapid metagenomic identification of viral pathogens in clinical samples by real-time nanopore sequencing analysis.Genome Med. 2015; 7: 99Crossref PubMed Scopus (262) Google Scholar, 22Leggett R.M. Alcon-Giner C. Heavens D. Caim S. Brook T.C. Kujawska M. Hoyles L. Clarke P. Hall L.J. Clark M.D. Rapid MinION metagenomic profiling of the preterm infant gut microbiota to aid in pathogen diagnostics.bioRxiv. 2017; : 180406Google Scholar, 23Quick J. Loman N.J. Duraffour S. Simpson J.T. Severi E. Cowley L. et al.Real-time, portable genome sequencing for Ebola surveillance.Nature. 2016; 530: 228-232Crossref PubMed Scopus (646) Google Scholar, 24Sanderson N.D. Street T.L. Foster D. Swann J. Atkins B.L. Brent A.J. McNally M.A. Oakley S. Taylor A. Peto T.E. Real-time analysis of nanopore-based metagenomic sequencing from orthopaedic device infection.bioRxiv. 2017; : 220616Google Scholar, 25Schmidt K. Mwaigwisya S. Crossman L.C. Doumith M. Munroe D. Pires C. Khan A.M. Woodford N. Saunders N.J. Wain J. O'Grady J. Livermore D.M. Identification of bacterial pathogens and antimicrobial resistance directly from clinical urines by nanopore-based metagenomic sequencing.J Antimicrob Chemother. 2017; 72: 104-114Crossref PubMed Scopus (144) Google Scholar, 26Quick J. Ashton P. Calus S. Chatt C. Gossain S. Hawker J. Nair S. Neal K. Nye K. Peters T. De Pinna E. Robinson E. Struthers K. Webber M. Catto A. Dallman T.J. Hawkey P. Loman N.J. Rapid draft sequencing and real-time nanopore sequencing in a hospital outbreak of Salmonella.Genome Biol. 2015; 16: 114Crossref PubMed Scopus (163) Google Scholar, 27Charalampous T. Richardson H. Kay G.L. Baldan R. Jeanes C. Rae D. Grundy S. Turner D.J. Wain J. Leggett R.M. Rapid diagnosis of lower respiratory infection using nanopore-based clinical metagenomics.bioRxiv. 2018; : 387548Google Scholar However, for sepsis diagnostics of patient samples, such a rapid real-time sequencing-based workflow has not been established yet. Because the nanopore sequencing platform sequences in real time with the potential to reduce time to diagnosis down to a few hours, a MinION-based workflow was established to determine its suitability for the diagnosis of sepsis in terms of throughput, sensitivity, specificity, and speed. Human data result from a secondary analysis of a subset of patients in a previously published, prospective observational clinical study,15Decker S.O. Sigl A. Grumaz C. Stevens P. Vainshtein Y. Zimmermann S. Weigand M.A. Hofer S. Sohn K. Brenner T. Immune-response patterns and next generation sequencing diagnostics for the detection of mycoses in patients with septic shock—results of a combined clinical and experimental investigation.Int J Mol Sci. 2017; 18: 1796Crossref Scopus (22) Google Scholar,16Grumaz S. Grumaz C. Vainshtein Y. Stevens P. Glanz K. Decker S.O. Hofer S. Weigand M.A. Brenner T. Sohn K. Enhanced performance of next-generation sequencing diagnostics compared with standard of care microbiological diagnostics in patients suffering from septic shock.Crit Care Med. 2019; 47: e394-e402Crossref PubMed Scopus (17) Google Scholar which was conducted in the surgical intensive care unit of Heidelberg University Hospital, Germany (German Clinical Trials Register: DRKS00005463). Study and control patients or their legal designees signed written informed consent. All study procedures were approved by the local ethics committee (Ethics Committee of the Medical Faculty of Heidelberg, Trial Code No. S-097/2013). Data result from a secondary analysis of a selection of four septic patients (S10, S16, S33, and S35) and three control surgical patients without any evidence for infection (P1, P2, and P13) participating in a previously published, prospective observational clinical study of our workgroup, which was conducted in the surgical intensive care unit of Heidelberg University Hospital, Germany (German Clinical Trials Register: DRKS00005463).15Decker S.O. Sigl A. Grumaz C. Stevens P. Vainshtein Y. Zimmermann S. Weigand M.A. Hofer S. Sohn K. Brenner T. Immune-response patterns and next generation sequencing diagnostics for the detection of mycoses in patients with septic shock—results of a combined clinical and experimental investigation.Int J Mol Sci. 2017; 18: 1796Crossref Scopus (22) Google Scholar,16Grumaz S. Grumaz C. Vainshtein Y. Stevens P. Glanz K. Decker S.O. Hofer S. Weigand M.A. Brenner T. Sohn K. Enhanced performance of next-generation sequencing diagnostics compared with standard of care microbiological diagnostics in patients suffering from septic shock.Crit Care Med. 2019; 47: e394-e402Crossref PubMed Scopus (17) Google Scholar Although the cfDNA of Patients S10, S33, and S35 was only analyzed at the onset of sepsis (T0), cfDNA of Patient S16 was analyzed also at later time points until 14 days after the onset with a total of five samples (T0 to T4: T1, day 1; T2, day 2; T3, day 7; and T4, day 14), resulting in a total of eight samples from four septic patients. Based on the primary and clinically validated analysis, this selection of septic samples showed either mono- or polymicrobial infections caused by a variety of bacteria, a virus, and a fungus with different burdens, providing a well-characterized and highly diversified template for this proof-of-concept study. In addition, cfDNA of three surgical patients without any evidence of infection were included as controls (P1 T0, P2 T0, and P13 T0), contributing to a total of 11 samples. Sepsis and control patients or their legal designees signed written informed consent. All study procedures were approved by the local ethics committee (Ethics Committee of the Medical Faculty of Heidelberg, Trial Code No. S-097/2013). S10: A 74–year-old male patient presented with a known pelvic liposarcoma, so that a hemicolectomy and a segmental resection of the small bowels with primary anastomosis were performed. Eight days later, the patient suffered from anastomotic leakage so that an additional surgical revision was necessary. This surgical revision included another small bowel resection with the need for an ileostomy as well as a stump by Hartmann. Another 4 days later, the patient developed septic shock due to an insufficiency of the ileostomy with subsequent diffuse peritonitis, so that another surgical revision was performed, in which the ileostomy was refixed and an extensive abdominal lavage was done. Empiric antibiotic therapy included imipenem/cilastatin, vancomycin, and caspofungin. Due to a positive culture with vancomycin-resistant enterococci from an abdominal smear, the antibiotic treatment regime was secondarily switched to tigecycline and ceftazidime. Seven days after the onset of septic shock, the antibiotic treatment regime had to be adapted once again due to a worsening of the patient's condition, so a combination therapy with meropenem, linezolid, and fluconazole was applied. However, the patient died 20 days after the onset of septic shock. S16: A 73–year-old male patient presented with a tumor of his bile duct with the need for a biliodigestive anastomosis as well as a cholecystectomy. Four days after the surgical procedure, the patient developed septic shock due to a duodenal ulcer perforation with the need for a total pancreatectomy, as well as a partial small bowel resection. Empiric antibiotic treatment was started with meropenem. Seven days after onset of septic shock, BC became positive for Enterococcus faecium. However, this finding did not result in a targeted anti-infective treatment with, for example, vancomycin or linezolid. Another 7 days later, BC revealed a growth of Candida albicans, so the pre-existing antifungal treatment with fluconazole was escalated to caspofungin. The patient died at day 31 after sepsis onset. S33: A 66–year-old female patient presented with septic shock due to a toxic megacolon with subsequent bowel perforation, so a subtotal colectomy was performed. Empiric antibiotic treatment was performed with imipenem/cilastatin. The patient recovered well and was able to be discharged home within the next 8 weeks. S35: Two days following a basalectomy of the right lung due to a penetrating liver abscess, a 71–year-old female patient presented with septic shock due to a perforation of the colon. Accordingly, a subtotal colectomy was performed, and an empiric anti-infective treatment with imipenem/cilastatin and fluconazole was established. Due to persisting septic shock, vancomycin was added 3 days after the onset of septic shock. At day 9 after the onset of septic shock, secretion of abdominal drainages revealed signs of fecal contamination, so another surgical revision was performed. However, another bowel leakage could not be identified. Afterward, the patient showed recurrent microbiological growth of Candida species in abdominal drainages, so an antifungal treatment with caspofungin was initiated. The patient recovered slowly, but was able to be discharged home 3 months after the onset of septic shock. All samples used in this study were already isolated within a previous study.16Grumaz S. Grumaz C. Vainshtein Y. Stevens P. Glanz K. Decker S.O. Hofer S. Weigand M.A. Brenner T. Sohn K. Enhanced performance of next-generation sequencing diagnostics compared with standard of care microbiological diagnostics in patients suffering from septic shock.Crit Care Med. 2019; 47: e394-e402Crossref PubMed Scopus (17) Google Scholar Briefly, for plasma preparation, blood samples were centrifuged at 1600 × g and 4°C for 10 minutes followed by a second centrifugation of the supernatant containing the plasma at 16,000 × g and 4°C for 10 minutes. Until further processing, plasma was stored at −80°C. cfDNA isolation was performed using the QIAsymphony and the QIAsymphony Circulating DNA kit (both from Qiagen, Hilden, Germany) according to the manufacturer's instructions with the following modification: Thawed plasma was centrifuged for 10 minutes at 16,000 × g and 4°C to remove precipitates. The elution of the cfDNA from beads was performed with 60 μL of molecular biology grade water (5 PRIME, Hamburg, Germany). The quantity and quality of cfDNA were assessed with the Qubit dsDNA HS Assay Kit by the Qubit 2.0 fluorometer (Thermo Fisher Scientific, Waltham, MA) and the DNF-474 High Sensitivity NGS Fragment Analysis Kit (1 bp to 6000 bp) using a Fragment Analyzer (Advanced Analytical Technologies, Heidelberg, Germany), respectively. Nanopore sequencing libraries were prepared using 5 ng of cfDNA using the Low Input genomic DNA by PCR/SQK-LWP001 Kit (Oxford Nanopore Technologies) according to the manufacturer's recommendations including the following exceptions: i) purification steps were performed with SPRIselect beads (Beckman Coulter, Brea, CA) and 80% ethanol; ii) adapter ligation was performed with the twofold amount of PCR adapter mix followed by a 1.0× volume SPRIselect beads purification; iii) PCR was performed with the HiFi HotStart DNA Polymerase (Roche, Mannheim, Germany) and the twofold amount of Whole Genome Primer under the following conditions: initial denaturation for 3 minutes at 95°C; 16 cycles of denaturation for 20 seconds at 98°C, annealing for 15 seconds at 56°C, and extension for 20 seconds at 72°C; and final extension for 1 minute at 72°C; iv) final PCR products were purified with 0.8× volume of SPRIselect beads; and v) for the one-directional read (1D) Adapter ligation, the twofold amount of the Rapid Adapter was used with 10 μL of library (approximately 200 fmole). For samples S16 T0 to T2, only a partially optimized protocol was used with original volumes of PCR adapter mix, Whole Genome Primer, and Rapid Adapter, and 20 cycles of PCR. Reduction of time was introduced by exchange of AMPure XP beads (Beckman Coulter) with SPRI beads and shortened extension time during PCR. Nanopore sequencing was performed according to the manufacturer's instructions using MinKNOW software version 1.6.11 (Oxford Nanopore Technologies) or higher on a MinION. Number of active pores ranged from 1074 to 1517 according to initial quality control prior sequencing start. Libraries from S16 T0 to S16 T1 were loaded on the SpotON Flow Cell R9.5, whereas all others were loaded on R9.4. All samples were sequenced for 48 hours, except S16 T1 for 44 hours. Illumina libraries were already sequenced within a previous study.16Grumaz S. Grumaz C. Vainshtein Y. Stevens P. Glanz K. Decker S.O. Hofer S. Weigand M.A. Brenner T. Sohn K. Enhanced performance of next-generation sequencing diagnostics compared with standard of care microbiological diagnostics in patients suffering from septic shock.Crit Care Med. 2019; 47: e394-e402Crossref PubMed Scopus (17) Google Scholar Library preparation and sequencing were performed as previously described14Grumaz S. Stevens P. Grumaz C. Decker S.O. Weigand M.A. Hofer S. Brenner T. von Haeseler A. Sohn K. Next-generation sequencing diagnostics of bacteremia in septic patients.Genome Med. 2016; 8: 73Crossref PubMed Scopus (124) Google Scholar from 1 ng of cfDNA using the Nextera XT library preparation kit (Illumina, San Diego, CA) with a Biomek FXP liquid handling robot (Beckman Coulter). Sequencing of the libraries was performed on a HiSeq2500 (Illumina), resulting in 33 million 100-bp single-end reads, on average, per sample. Raw reads (FAST5) were acquired and basecalled using MinKNOW version 1.6.11 software or higher according to a protocol for R9.4 or R9.5 flowcells.28Ip C.L. Loose M. Tyson J.R. de Cesare M. Brown B.L. Jain M. Leggett R.M. Eccles D.A. Zalunin V. Urban J.M. MinION Analysis and Reference Consortium: phase 1 data release and analysis.F1000Res. 2015; 4: 1075Crossref PubMed Scopus (147) Google Scholar,29Lu H. Giordano F. Ning Z. Oxford nanopore MinION sequencing and genome assembly.Genomics Proteomics Bioinformatics. 2016; 14: 265-279Crossref PubMed Scopus (278) Google Scholar Base-called MinION reads were cleared from potential adapter contamination using Porechop software version 0.2.3 (GitHub, https://github.com/rrwick/Porechop, last accessed August 14, 2019), based on SeqAn library.30Reinert K. Dadi T.H. Ehrhardt M. Hauswedell H. Mehringer S. Rahn R. Kim J. Pockrandt C. Winkler J. Siragusa E. Urgese G. Weese D. The SeqAn C++ template library for efficient sequence analysis: a resource for programmers.J Biotechnol. 2017; 261: 157-168Crossref PubMed Scopus (34) Google Scholar In addition, all reads shorter than 50 bp were filtered out, and all reads longer than 2000 bp were trimmed to this maximum length using BBDuk in a single-end mode from the BBMap package version 34.41 (SourceForge, https://sourceforge.net/projects/bbmap, last accessed August 14, 2019). Also using BBDuk, all reads below a Phred score of 10 were skipped. The quality of raw and trimmed reads was checked using the NanoPlot version 1.14.1 software package.31De Coster W. D'Hert S. Schultz D.T. Cruts M. Van Broeckhoven C. NanoPack: visualizing and processing long-read sequencing data.Bioinformatics. 2018; 34: 2666-2669Crossref PubMed Scopus (354) Google Scholar To account for low-complexity sequence regions, PRINSEQ-lite version 0.20.4 software program (SourceForge, http://prinseq.sourceforge.net, last accessed August 14, 2019) with -lc_method dust -lc_threshold 7 options was used.32Schmieder R. Edwards R. Quality control and preprocessing of metagenomic datasets.Bioinformatics. 2011; 27: 863-864Crossref PubMed Scopus (2533) Google Scholar The DUST approach is adapted from the algorithm used to mask low-complexity regions during BLAST search preprocessing.33Morgulis A. Gertz E.M. Schaffer A.A. Agarwala R. A fast and symmetric DUST implementation to mask low-complexity DNA sequences.J Comput Biol. 2006; 13: 1028-1040Crossref PubMed Scopus (242) Google Scholar Each sample was mapped to a human reference genome (GRCh38.80) using Burrows-Wheeler Aligner version bwa-0.7.17 –r1194 (SourceForge, https://sourceforge.net/projects/bio-bwa/files, last accessed August 14, 2019) with BWA-MEM algorithm and -x ont2d flag34Li H. Durbin R. Fast and accurate long-read alignment with Burrows-Wheeler transform.Bioinformatics. 2010; 26: 589-595Crossref PubMed Scopus (5669) Google Scholar; output BAM file was indexed and sorted with SAMtools version 0.1.18 to further control percent identity to the reference.35Li H. A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data.Bioinformatics. 2011; 27: 2987-2993Crossref PubMed Scopus (2271) Google Scholar,36Li H. Handsaker B. Wysoker A. Fennell T. Ruan J. Homer N. Marth G. Abecasis G. Durbin R. 1000 Genome Project Data Processing SubgroupThe Sequence Alignment/Map format and SAMtools.Bioinformatics. 2009; 25: 2078-2079Crossref PubMed Scopus (24663) Google Scholar As described in previous work,14Grumaz S. Stevens P. Grumaz C. Decker S.O. Weigand M.A. Hofer S. Brenner T. von Haeseler A. Sohn K. Next-generation sequencing diagnostics of bacteremia in septic patients.Genome Med. 2016; 8: 73Crossref PubMed Scopus (124) Google Scholar Kraken version 1.0 (Johns Hopkins University Center for Computational Biology, http://ccb.jhu.edu/software/kraken, last accessed August 14, 2019) was then used to assign unmapped reads to taxons using the National Center for Biotechnology Information Reference Sequence (RefSeq) database (release version 68; https://www.ncbi.nlm.nih.gov/refseq) comprising 35,749 bacterial and 4340 viral genomes complemented by 16 selected fungal genomes. Additionally, unmapped reads were also mapped with BWA-MEM to a custom genome collection derived from 74 sepsis-relevant species that were identified in a previous study and further labeled as SIQ-database (SIQ-db) in the paper.16Grumaz S. Grumaz C. Vainshtein Y. Stevens P. Glanz K. Decker S.O. Hofer S.