False Positive Assignments Research Articles

Characterization and bioinformatic filtering of ambient gRNAs in single-cell CRISPR screens using CLEANSER.

Recent technological developments in single-cell RNA-seq CRISPR screens enable high-throughput investigation of the genome. Through transduction of a gRNA library to a cell population followed by transcriptomic profiling by scRNA-seq, it is possible to characterize the effects of thousands of genomic perturbations on global gene expression. A major source of noise in scRNA-seq CRISPR screens are ambient gRNAs, which are contaminating gRNAs that likely originate from other cells. If not properly filtered, ambient gRNAs can result in an excess of false positive gRNA assignments. Here, we utilize CRISPR barnyard assays to characterize ambient gRNA noise in single-cell CRISPR screens. We use these datasets to develop and train CLEANSER, a mixture model that identifies and filters ambient gRNA noise. This model takes advantage of the bimodal distribution between native and ambient gRNAs and includes both gRNA and cell-specific normalization parameters, correcting for confounding technical factors that affect individual gRNAs and cells. The output of CLEANSER is the probability that a gRNA-cell assignment is in the native distribution over the ambient distribution. We find that ambient gRNA filtering methods impact differential gene expression analysis outcomes and that CLEANSER outperforms alternate approaches by increasing gRNA-cell assignment accuracy.

Development and validation of novel SNP markers for the rapid identification of natural hybrids of the 11 closely related pufferfish species (Takifugu spp.) distributed in Japan

Accurate identification of interspecific hybrids is vital for breeding programs in hybridizing taxa, as well as for the use of hybrids in aquaculture and fisheries. The use of species-specific single nucleotide polymorphisms (SNPs) has become the standard for the molecular identification of hybrids. However, it remains challenging to use these SNPs to identify hybrids involving multiple candidate parental species. Here, we developed and validated SNP markers for the identification of hybrids of the 11 closely related pufferfish species (genus Takifugu) distributed in Japan; these species have different aquacultural characteristics, such as growth rate, physiology, behavior, and toxicity. In total, 1232 species-specific SNPs were identified through double-digest restriction-site-associated DNA sequencing. The species had 11 (T. flavipterus) to 295 (T. chrysops) species-specific SNPs. We validated 33 genome-wide species-specific SNPs using TaqMan® genotyping assays for the 11 species and their hybrids, which were identified using amplified fragment length polymorphism analyses. No false negative or positive assignments were observed in 77 and 30 specimens of pure species and F1 hybrids, respectively. This marker combination allows rapid and reliable identification of natural hybrids, which will be useful for pufferfish aquaculture and will enhance our understanding of the natural hybridization between various Takifugu species.

Evolution of Magnetic Resonance Imaging Lesions at the Sacroiliac Joints During and After Pregnancy by Serial Magnetic Resonance Imaging From Gestational Week Twenty to Twelve Months Postpartum.

Sacroiliac (SI) joint magnetic resonance imaging (MRI) findings simulating sacroiliitis related to axial spondyloarthritis (SpA) may occur in women before and after birth. This study was undertaken to explore the prevalence, evolution, and topography of SI joint MRI lesions in pregnant and postpartum women. A prospective cohort study included 103 first-time mothers who underwent up to 5 serial SI joint MRI between gestational week 20 and 12 months postpartum. After calibration, 3 assessors independently evaluated bone marrow edema (BME), including sacroiliitis according to the Assessment of SpondyloArthritis international Society (ASAS), as well as structural lesions, using the Spondyloarthritis Research Consortium of Canada (SPARCC) and a novel 2-plane assessment method. BME was frequent both during pregnancy and the postpartum period, peaking at 3 months postpartum with a prevalence of 69% (SPARCC) and 80% (2-plane method), but still present in 54% (SPARCC) and 58% (2-plane method) of subjects at 12 months postpartum. At 12 months postpartum, sacroiliitis according to the current ASAS definition was met in 41%, while 21% and 14% of women fulfilled the newly proposed ASAS MRI thresholds for active and structural SI joint lesions, respectively. BME clustered in the anterior middle joint portions at all time points, and ligamentous BME was rare. At 12 months postpartum, SPARCC erosion scores ≥3 (ASAS threshold) were observed in only 2.8% of women. At 12 months postpartum, 41% of women met the current ASAS sacroiliitis definition, which may result in false-positive assignments of axial SpA diagnosis in postpartum women with back pain. The topographical BME distribution and virtually absent erosions (ASAS threshold) at 12 months postpartum may help discriminate postpartum strain-related conditions from axial SpA-related sacroiliitis.

A data processing pipeline for petroleomics based on liquid chromatography-high resolution mass spectrometry

Online liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS) has attracted much attention in the molecular characterization of crude oil. Neither open access nor commercially available petroleomics tools were developed specifically to process LC-HRMS data. Here, a novel data processing pipeline was specifically designed for LC-HRMS-based petroleomics data. A customizable formula database was established deriving from the detected sample, which could avoid the interference caused by a large number of redundant molecules in a conventionally theoretical molecular database. Molecular formula candidates were assigned by the formula database using a low noise threshold, and false-positive assignments were eliminated by the chromatographic retention behaviors. Multi-dimensional information was obtained, including heteroatom class, double bond equivalent (DBE), carbon number, retention time, and MS/MS spectra. The developed method was compared with a popular petroleomics software, similar relative abundance class distribution was obtained, and much more formulas of low abundant components were uniquely extracted by the developed method. Finally, it was applied to reveal variation between feed and product oils in hydrodenitrogenation. Significantly compositional and structural differences were revealed. The developed method provides a useful pipeline for molecular data mining of petroleum samples.

Sterols and sterol ratios to trace fecal contamination: pitfalls and potential solutions.

Fecal pollution in surface waters is a major threat to recreational and drinking water resources, with Escherichia coli being a primary concern. The best way to mitigate fecal pollutant loading is to identify the sources and tailor remediation strategies to reduce loading. Tracking E. coli back to its source is notoriously difficult in a mixed-use watershed where input from humans, wildlife, and livestock all contribute to E. coli loading. One proposed tracking method for E. coli contamination is the use of fecal sterols and sterol ratios. This study uses fecal sterol data published globally to assess how well sterol compositions for different species clusters along with the effectiveness of sterol ratios as tracking tools. Hierarchical cluster analysis produces stronger clusters based on sterol ratios than raw sterol concentration, but the global dataset results in clustering of the same species in different levels. The accuracy of the sterol ratios was also compared to understand the rate of false negatives and false positive assignments. Overall, these ratios did not have a high success rate for determining the correct source, which was also reflected in the poor clustering trends observed. Establishing local end-member sterol profiles is essential when using sterol signatures to unravel fecal loading.

MCycDB: A curated database for comprehensively profiling methane cycling processes of environmental microbiomes.

Methane is a critical greenhouse gas with significant impacts on environmental and global change. However, CH4 cycling processes and coupling mechanisms with the biogeochemical cycling of carbon, nitrogen, sulfur and metals in the environment remain elusive. To fill such knowledge gaps, we constructed a manually curated methane cycling database (MCycDB) for comprehensive and accurate analysis of methane cycling microbial communities. MCycDB contains 298methane cycling gene families covering 10methane metabolism pathways with 610,208 representative sequences, and associated reference sequences from the NCBI RefSeq database with 48 phyla and 2,197genera, and five phyla and 100genera for bacteria and archaea, respectively. Also, homologous groups from public orthology databases were identified and included in MCycDB to reduce false positive assignments. We applied MCycDB to profile methane cycling gene families and associated taxonomic groups from various environments. Gene families involved in methanogenesis were abundant in hot spring sediment and less abundant in freshwater, whereas the ones involved in aerobic oxidation of methane were abundant in permafrost and peatland. This study demonstrates that MCycDB is a useful tool for studying microbially-driven methane cycling processes with high specificity, coverage and accuracy.

LncRNA functional annotation with improved false discovery rate achieved by disease associations

The long non‐coding RNAs (lncRNAs) play critical roles in various biological processes and are associated with many diseases. Functional annotation of lncRNAs in diseases attracts great attention in understanding their etiology. However, the traditional co-expression-based analysis usually produces a significant number of false positive function assignments. It is thus crucial to develop a new approach to obtain lower false discovery rate for functional annotation of lncRNAs. Here, a novel strategy termed DAnet which combining disease associations with cis-regulatory network between lncRNAs and neighboring protein-coding genes was developed, and the performance of DAnet was systematically compared with that of the traditional differential expression-based approach. Based on a gold standard analysis of the experimentally validated lncRNAs, the proposed strategy was found to perform better in identifying the experimentally validated lncRNAs compared with the other method. Moreover, the majority of biological pathways (40%∼100%) identified by DAnet were reported to be associated with the studied diseases. In sum, the DAnet is expected to be used to identify the function of specific lncRNAs in a particular disease or multiple diseases.

A Streamlined Data Analysis Pipeline for the Identification of Sites of Citrullination.

Citrullination is an enzyme-catalyzed post-translational modification (PTM) that is essential for a host of biological processes, including gene regulation, programmed cell death, and organ development. While this PTM is required for normal cellular functions, aberrant citrullination is a hallmark of autoimmune disorders as well as cancer. Although aberrant citrullination is linked to human pathology, the exact role of citrullination in disease remains poorly characterized, in part because of the challenges associated with identifying the specific arginine residues that are citrullinated. Tandem mass spectrometry is the most precise method for uncovering sites of citrullination; however, due to the small mass shift (+0.984 Da) that results from citrullination, current database search algorithms commonly misannotate spectra, leading to a high number of false-positive assignments. To address this challenge, we developed an automated workflow to rigorously and rapidly mine proteomic data to unambiguously identify the sites of citrullination from complex peptide mixtures. The crux of this streamlined workflow is the ionFinder software program, which classifies citrullination sites with high confidence on the basis of the presence of diagnostic fragment ions. These diagnostic ions include the neutral loss of isocyanic acid, which is a dissociative event that is unique to citrulline residues. Using the ionFinder program, we have mapped the sites of autocitrullination on purified protein arginine deiminases (PAD1-4) and mapped the global citrullinome in a PAD2-overexpressing cell line. The ionFinder algorithm is a highly versatile, user-friendly, and open-source program that is agnostic to the type of instrument and mode of fragmentation that are used.

VB12Path for Accurate Metagenomic Profiling of Microbially Driven Cobalamin Synthesis Pathways.

ABSTRACTCobalamin (vitamin B12; VB12) is an indispensable nutrient for all living entities in the Earth’s biosphere and plays a vital role in both natural and host environments. Currently in the metagenomic era, gene families of interest are extracted and analyzed based on functional profiles by searching shotgun metagenomes against public databases. However, critical issues exist in applying public databases for specific processes such as VB12 biosynthesis pathways. We developed a curated functional gene database termed VB12Path for accurate metagenomic profiling of VB12 biosynthesis gene families of microbial communities in complex environments. VB12Path contains a total of 60 VB12 synthesis gene families, 287,731 sequences, and 21,154 homology groups, and it aims to provide accurate functional and taxonomic profiles of VB12 synthesis pathways for shotgun metagenomes and minimize false-positive assignments. VB12Path was applied to characterize cobalamin biosynthesis gene families in human intestines and marine environments. The results demonstrated that ocean and human intestine had dramatically different VB12 synthesis processes and that gene families belonging to salvage and remodeling pathway dominated human intestine but were lowest in the ocean ecosystem. VB12Path is expected to be a useful tool to study cobalamin biosynthesis processes via shotgun metagenome sequencing in both environmental and human microbiome research.IMPORTANCE Vitamin B12 (VB12) is an indispensable nutrient for all living entities in the world but can only be synthesized by a small subset of prokaryotes. Therefore, this small subset of prokaryotes controls ecosystem stability and host health to some extent. However, critical accuracy and comprehensiveness issues exist in applying public databases to profile VB12 synthetic gene families and taxonomic groups in complex metagenomes. In this study, we developed a curated functional gene database termed VB12Path for accurate metagenomic profiling of VB12 communities in complex environments. VB12Path is expected to serve as a valuable tool to uncover the hidden microbial communities producing this precious nutrient on Earth.

Identification of spliced peptides in pancreatic islets uncovers errors leading to false assignments

Proteasomal spliced peptides (PSPs) have been identified in the class I major histocompatibility complex (MHC) peptidomes of several tumors and have emerged as novel neoantigens that can stimulate highly specific T cells. Much debate has surrounded the percentage of PSPs in the immunopeptidome; reported numbers have ranged from <1-5% to 12-45%. Recently, our laboratory demonstrated in nonobese diabetic (NOD) mice that hybrid insulin peptides (HIPs), a special class of spliced peptides, are formed during insulin granule degradation in crinosomes of the pancreatic β cells and that modified peptides comprised a significant source of false positive HIP assignments. Herein, this study is extended to crinosomes isolated from other mouse strains and to two recent MHC class I studies, to see if modified peptides explained discrepancies in reported percentages of PSPs. This analysis revealed that both MHC-I peptidomes contained many spectra erroneously assigned as PSPs. While many false positive PSPs did arise from modified peptides, others arose from probable data processing errors. Thus, the reported numbers of PSPs in the literature are likely elevated due to errors associated with data processing and analysis.

Modelling the utility of group testing for public health surveillance.

In epidemic or pandemic situations, resources for testing the infection status of individuals may be scarce. Although group testing can help to significantly increase testing capabilities, the (repeated) testing of entire populations can exceed the resources of any country. We thus propose an extension of the theory of group testing that takes into account the fact that definitely specifying the infection status of each individual is impossible. Our theory builds on assigning to each individual an infection status (healthy/infected), as well as an associated cost function for erroneous assignments. This cost function is versatile, e.g., it could take into account that false negative assignments are worse than false positive assignments and that false assignments in critical areas, such as health care workers, are more severe than in the general population. Based on this model, we study the optimal use of a limited number of tests to minimize the expected cost. More specifically, we utilize information-theoretic methods to give a lower bound on the expected cost and describe simple strategies that can significantly reduce the expected cost over currently known strategies. A detailed example is provided to illustrate our theory.

SCycDB: A curated functional gene database for metagenomic profiling of sulphur cycling pathways

AbstractMicroorganisms play important roles in the biogeochemical cycling of sulphur (S), an essential element in the Earth's biosphere. Shotgun metagenome sequencing has opened a new avenue to advance our understanding of S cycling microbial communities. However, accurate metagenomic profiling of S cycling microbial communities remains technically challenging, mainly due to low coverage and inaccurate definition of S cycling gene families in public orthology databases. Here we developed a manually curated S cycling database (SCycDB) to profile S cycling functional genes and taxonomic groups for shotgun metagenomes. The developed SCycDB contains 207 gene families and 585,055 representative sequences affiliated with 52 phyla and 2684 genera of bacteria/archaea, and 20,761 homologous orthology groups were also included to reduce false positive sequence assignments. SCycDB was applied for functional and taxonomic analysis of S cycling microbial communities from four habitats (freshwater, hot spring, marine sediment and soil). Gene families and microorganisms involved in S reduction were abundant in the marine sediment, while those of S oxidation and dimethylsulphoniopropionate transformation were abundant in the soil. SCycDB is expected to be a useful tool for fast and accurate metagenomic analysis of S cycling microbial communities in the environment.

Best match graphs and reconciliation of gene trees with species trees

A wide variety of problems in computational biology, most notably the assessment of orthology, are solved with the help of reciprocal best matches. Using an evolutionary definition of best matches that captures the intuition behind the concept we clarify rigorously the relationships between reciprocal best matches, orthology, and evolutionary events under the assumption of duplication/loss scenarios. We show that the orthology graph is a subgraph of the reciprocal best match graph (RBMG). We furthermore give conditions under which an RBMG that is a cograph identifies the correct orthlogy relation. Using computer simulations we find that most false positive orthology assignments can be identified as so-called good quartets—and thus corrected—in the absence of horizontal transfer. Horizontal transfer, however, may introduce also false-negative orthology assignments.

Raman spectroscopy for forensic bloodstain identification: Method validation vs. environmental interferences

The identification of body fluid (BF) stains at a crime scene is an important step of a crime scene investigation since traces of BFs are typically the main source of DNA evidence. Raman spectroscopy is a new emerging method for identifying and differentiating between BF stains. As any new analytical method developed for identification and differentiation of classes of materials, it needs to be evaluated for potential environmental interferences (EIs). This study specifically focused on proving that Raman spectroscopy is not susceptible to false positive (FP) assignments from EIs specific to blood. Twenty-four EI substances that may be misclassified as blood due to either their appearance, or known to provide a FP result with forensic crime scene tests for blood identification were analyzed in this study. The spectra of the twenty-four substances were tested against a recently developed support vector machines discriminant analysis (SVMDA) classification model used for BF differentiation. The SVMDA model identified all substances as not peripheral blood, but some of the substances were predicted as one of the other four BFs. Therefore, random forests (RF) analysis was employed to build a new, more robust, classification method to overcome this limitation. Using this approach, complete separation was achieved based on a classification probability threshold of 70%. This method can also be used in the future to test potential EI substances specific to other common BFs.

Raman microspectroscopic mapping as a tool for detection of gunshot residue on adhesive tape.

Our research group previously reported a novel method for the detection of gunshot residue (GSR) via tape lifting combined with Raman microspectroscopic mapping and multivariate analysis. This initial study achieved proof of concept for this approach. Here, we report validation studies which investigate the reproducibility/ruggedness and specificity of the approach. Raman mapping for GSR detection on adhesive tape was performed on an independent Raman microscope, not used to generate the training set. These independent spectra were classified against the original training dataset using support vector machine discriminant analysis (SVM-DA). The resulting classification rates of 100% illustrate the reproducibility of the technique, its independence upon a specific instrument and provide an external validation for the approach. Additionally, the same procedure for GSR collection (tape lifting) was performed to collect samples from environmental sources, which could potentially provide false-positive assignments for current GSR analysis techniques. Thus, particles associated with automotive mechanics were collected. Automotive brake and tire materials are often composed of the heavy metals lead, barium, and antimony, which are the key elements targeted by current GSR detection technique. It was determined that Raman spectroscopic analysis was not susceptible to misclassifications from these samples. Results from these validation experiments illustrate the great potential of Raman microspectroscopic mapping used with tape lifting as a viable complimentary tool to current methodologies for GSR detection. Furthermore, current methodologies are not well-developed for automated organic GSR detection. Illustrated here, Raman microscoptrosocpic mapping has the potential for the automatic identification of organic GSR. Graphical abstract ᅟ.

ProGeM: a framework for the prioritization of candidate causal genes at molecular quantitative trait loci.

Quantitative trait locus (QTL) mapping of molecular phenotypes such as metabolites, lipids and proteins through genome-wide association studies represents a powerful means of highlighting molecular mechanisms relevant to human diseases. However, a major challenge of this approach is to identify the causal gene(s) at the observed QTLs. Here, we present a framework for the ‘Prioritization of candidate causal Genes at Molecular QTLs’ (ProGeM), which incorporates biological domain-specific annotation data alongside genome annotation data from multiple repositories. We assessed the performance of ProGeM using a reference set of 227 previously reported and extensively curated metabolite QTLs. For 98% of these loci, the expert-curated gene was one of the candidate causal genes prioritized by ProGeM. Benchmarking analyses revealed that 69% of the causal candidates were nearest to the sentinel variant at the investigated molecular QTLs, indicating that genomic proximity is the most reliable indicator of ‘true positive’ causal genes. In contrast, cis-gene expression QTL data led to three false positive candidate causal gene assignments for every one true positive assignment. We provide evidence that these conclusions also apply to other molecular phenotypes, suggesting that ProGeM is a powerful and versatile tool for annotating molecular QTLs. ProGeM is freely available via GitHub.

NCycDB: a curated integrative database for fast and accurate metagenomic profiling of nitrogen cycling genes.

The nitrogen (N) cycle is a collection of important biogeochemical pathways in the Earth ecosystem and has gained extensive foci in ecology and environmental studies. Currently, shotgun metagenome sequencing has been widely applied to explore gene families responsible for N cycle processes. However, there are problems in applying publically available orthology databases to profile N cycle gene families in shotgun metagenomes, such as inefficient database searching, unspecific orthology groups and low coverage of N cycle genes and/or gene (sub)families. To solve these issues, this study built a manually curated integrative database (NCycDB) for fast and accurate profiling of N cycle gene (sub)families from shotgun metagenome sequencing data. NCycDB contains a total of 68 gene (sub)families and covers eight N cycle processes with 84 759 and 219 146 representative sequences at 95 and 100% identity cutoffs, respectively. We also identified 1958 homologous orthology groups and included corresponding sequences in the database to avoid false positive assignments due to 'small database' issues. We applied NCycDB to characterize N cycle gene (sub)families in 52 shotgun metagenomes from the Global Ocean Sampling expedition. Further analysis showed that the structure and composition of N cycle gene families were most strongly correlated with latitude and temperature. NCycDB is expected to facilitate N cycle studies via shotgun metagenome sequencing approaches in various environments. The framework developed in this study can be served as a good reference to build similar knowledge-based functional gene databases in various processes and pathways. NCycDB database files are available at https://github.com/qichao1984/NCyc. Supplementary data are available at Bioinformatics online.

Principles of Systems Biology, No. 31.

This month: selected work from the 2018 RECOMB meeting, organized by Ecole Polytechnique and held last April in Paris.

Fast Genome-Wide Functional Annotation through Orthology Assignment by eggNOG-Mapper.

Orthology assignment is ideally suited for functional inference. However, because predicting orthology is computationally intensive at large scale, and most pipelines are relatively inaccessible (e.g., new assignments only available through database updates), less precise homology-based functional transfer is still the default for (meta-)genome annotation. We, therefore, developed eggNOG-mapper, a tool for functional annotation of large sets of sequences based on fast orthology assignments using precomputed clusters and phylogenies from the eggNOG database. To validate our method, we benchmarked Gene Ontology (GO) predictions against two widely used homology-based approaches: BLAST and InterProScan. Orthology filters applied to BLAST results reduced the rate of false positive assignments by 11%, and increased the ratio of experimentally validated terms recovered over all terms assigned per protein by 15%. Compared with InterProScan, eggNOG-mapper achieved similar proteome coverage and precision while predicting, on average, 41 more terms per protein and increasing the rate of experimentally validated terms recovered over total term assignments per protein by 35%. EggNOG-mapper predictions scored within the top-5 methods in the three GO categories using the CAFA2 NK-partial benchmark. Finally, we evaluated eggNOG-mapper for functional annotation of metagenomics data, yielding better performance than interProScan. eggNOG-mapper runs ∼15× faster than BLAST and at least 2.5× faster than InterProScan. The tool is available standalone and as an online service at http://eggnog-mapper.embl.de.

Forensic Hair Differentiation Using Attenuated Total Reflection Fourier Transform Infrared (ATR FT-IR) Spectroscopy.

Hair and fibers are common forms of trace evidence found at crime scenes. The current methodology of microscopic examination of potential hair evidence is absent of statistical measures of performance, and examiner results for identification can be subjective. Here, attenuated total reflection (ATR) Fourier transform-infrared (FT-IR) spectroscopy was used to analyze synthetic fibers and natural hairs of human, cat, and dog origin. Chemometric analysis was used to differentiate hair spectra from the three different species, and to predict unknown hairs to their proper species class, with a high degree of certainty. A species-specific partial least squares discriminant analysis (PLSDA) model was constructed to discriminate human hair from cat and dog hairs. This model was successful in distinguishing between the three classes and, more importantly, all human samples were correctly predicted as human. An external validation resulted in zero false positive and false negative assignments for the human class. From a forensic perspective, this technique would be complementary to microscopic hair examination, and in no way replace it. As such, this methodology is able to provide a statistical measure of confidence to the identification of a sample of human, cat, and dog hair, which was called for in the 2009 National Academy of Sciences report. More importantly, this approach is non-destructive, rapid, can provide reliable results, and requires no sample preparation, making it of ample importance to the field of forensic science.