Reducing Sanger confirmation testing through false positive prediction algorithms.
Ontology highlight
ABSTRACT: Purpose
Clinical genome sequencing (cGS) followed by orthogonal confirmatory testing is standard practice. While orthogonal testing significantly improves specificity, it also results in increased turnaround time and cost of testing. The purpose of this study is to evaluate machine learning models trained to identify false positive variants in cGS data to reduce the need for orthogonal testing.Methods
We sequenced five reference human genome samples characterized by the Genome in a Bottle Consortium (GIAB) and compared the results with an established set of variants for each genome referred to as a truth set. We then trained machine learning models to identify variants that were labeled as false positives.Results
After training, the models identified 99.5% of the false positive heterozygous single-nucleotide variants (SNVs) and heterozygous insertions/deletions variants (indels) while reducing confirmatory testing of nonactionable, nonprimary SNVs by 85% and indels by 75%. Employing the algorithm in clinical practice reduced overall orthogonal testing using dideoxynucleotide (Sanger) sequencing by 71%.Conclusion
Our results indicate that a low false positive call rate can be maintained while significantly reducing the need for confirmatory testing. The framework that generated our models and results is publicly available at https://github.com/HudsonAlpha/STEVE .
SUBMITTER: Holt JM
PROVIDER: S-EPMC8257489 | biostudies-literature | 2021 Jul
REPOSITORIES: biostudies-literature
Publications
Reducing Sanger confirmation testing through false positive prediction algorithms.
Holt James M JM Kelly Melissa M Sundlof Brett B Nakouzi Ghunwa G Bick David D Lyon Elaine E
Genetics in medicine : official journal of the American College of Medical Genetics 20210325 7
<h4>Purpose</h4>Clinical genome sequencing (cGS) followed by orthogonal confirmatory testing is standard practice. While orthogonal testing significantly improves specificity, it also results in increased turnaround time and cost of testing. The purpose of this study is to evaluate machine learning models trained to identify false positive variants in cGS data to reduce the need for orthogonal testing.<h4>Methods</h4>We sequenced five reference human genome samples characterized by the Genome in ...[more]
Similar Datasets
Project description:PurposeThere is increasing demand from the public for direct-to-consumer (DTC) genetic tests, and the US Food and Drug Administration limits the type of health-related claims DTC tests can market. Some DTC companies provide raw genotyping data to customers if requested, and these raw data may include variants occurring in genes recommended by the American College of Medical Genetics and Genomics to be reported as incidental/secondary findings. The purpose of this study was to review the outcome of requests for clinical confirmation of DTC results that were received by our laboratory and to analyze variant classification concordance.MethodsWe identified 49 patient samples received for further testing that had previously identified genetic variants reported in DTC raw data. For each case identified, information pertaining to the outcome of clinical confirmation testing as well as classification of the DTC variant was collected and analyzed.ResultsOur analyses indicated that 40% of variants in a variety of genes reported in DTC raw data were false positives. In addition, some variants designated with the "increased risk" classification in DTC raw data or by a third-party interpretation service were classified as benign at Ambry Genetics as well as several other clinical laboratories, and are noted to be common variants in publicly available population frequency databases.ConclusionOur results demonstrate the importance of confirming DTC raw data variants in a clinical laboratory that is well versed in both complex variant detection and classification.
| S-EPMC6301953 | biostudies-literature
Project description:Newborn screening (NBS) for inborn metabolic disorders is a highly successful public health program that by design is accompanied by false-positive results. Here we trained a Random Forest machine learning classifier on screening data to improve prediction of true and false positives. Data included 39 metabolic analytes detected by tandem mass spectrometry and clinical variables such as gestational age and birth weight. Analytical performance was evaluated for a cohort of 2777 screen positives reported by the California NBS program, which consisted of 235 confirmed cases and 2542 false positives for one of four disorders: glutaric acidemia type 1 (GA-1), methylmalonic acidemia (MMA), ornithine transcarbamylase deficiency (OTCD), and very long-chain acyl-CoA dehydrogenase deficiency (VLCADD). Without changing the sensitivity to detect these disorders in screening, Random Forest-based analysis of all metabolites reduced the number of false positives for GA-1 by 89%, for MMA by 45%, for OTCD by 98%, and for VLCADD by 2%. All primary disease markers and previously reported analytes such as methionine for MMA and OTCD were among the top-ranked analytes. Random Forest's ability to classify GA-1 false positives was found similar to results obtained using Clinical Laboratory Integrated Reports (CLIR). We developed an online Random Forest tool for interpretive analysis of increasingly complex data from newborn screening.
| S-EPMC7080200 | biostudies-literature
Project description:BackgroundNext-generation sequencing pipelines often perform error correction as a preprocessing step to obtain cleaned input data. State-of-the-art error correction programs are able to reliably detect and correct the majority of sequencing errors. However, they also introduce new errors by making false-positive corrections. These correction mistakes can have negative impact on downstream analysis, such as k-mer statistics, de-novo assembly, and variant calling. This motivates the need for more precise error correction tools.ResultsWe present CARE 2.0, a context-aware read error correction tool based on multiple sequence alignment targeting Illumina datasets. In addition to a number of newly introduced optimizations its most significant change is the replacement of CARE 1.0's hand-crafted correction conditions with a novel classifier based on random decision forests trained on Illumina data. This results in up to two orders-of-magnitude fewer false-positive corrections compared to other state-of-the-art error correction software. At the same time, CARE 2.0 is able to achieve high numbers of true-positive corrections comparable to its competitors. On a simulated full human dataset with 914M reads CARE 2.0 generates only 1.2M false positives (FPs) (and 801.4M true positives (TPs)) at a highly competitive runtime while the best corrections achieved by other state-of-the-art tools contain at least 3.9M FPs and at most 814.5M TPs. Better de-novo assembly and improved k-mer analysis show the applicability of CARE 2.0 to real-world data.ConclusionFalse-positive corrections can negatively influence down-stream analysis. The precision of CARE 2.0 greatly reduces the number of those corrections compared to other state-of-the-art programs including BFC, Karect, Musket, Bcool, SGA, and Lighter. Thus, higher-quality datasets are produced which improve k-mer analysis and de-novo assembly in real-world datasets which demonstrates the applicability of machine learning techniques in the context of sequencing read error correction. CARE 2.0 is written in C++/CUDA for Linux systems and can be run on the CPU as well as on CUDA-enabled GPUs. It is available at https://github.com/fkallen/CARE .
| S-EPMC9195321 | biostudies-literature
Project description:BackgroundImplantable loop recorders (ILRs) are increasingly placed for arrhythmia detection. However, historically, ≈75% of ILR alerts are false positives, requiring significant time and effort for adjudication. The LINQII and LUX-Dx are remotely reprogrammable ILRs with dual-stage algorithms using artificial intelligence to reduce false positives, but their utility in routine clinical practice has not been studied.Methods and resultsWe identified patients with the LINQII and LUX-Dx who were monitored by the Veterans Affairs National Cardiac Device Surveillance Program between March and June 2022. ILR programming was customized on the basis of implant indication. All alerts and every 90-day scheduled transmissions were manually reviewed. ILRs were remotely reprogrammed, as appropriate, after false-positive alerts or 2 consecutive same-type alerts, unless there was ongoing clinical need for that alert. Outcomes were total number of transmissions and false positives. We performed medical record review to determine if patients experienced any adverse clinical events, including hospitalization and mortality. Among 117 LINQII patients, there were 239 total alerts, 43 (18.0%) of which were false positives. Among 105 LUX-Dx patients, there were 300 total alerts, 115 (38.3%) of which were false positives. LINQIIs were reprogrammed 22 times, resulting in a decrease in median alerts/day from 0.13 to 0.03. LUX-Dx ILRs were reprogrammed 52 times, resulting in a decrease from 0.15 to 0.01 median alerts/day. There were no adverse clinical events that could have been identified by superior or earlier arrhythmia detection.ConclusionsILRs with artificial intelligence algorithms and remote reprogramming ability are associated with reduced alert burden because of higher true-positive rates than prior ILRs, without missing potentially consequential arrhythmias.
| S-EPMC10944033 | biostudies-literature
Project description:Coronavirus disease 2019 (COVID-19) has become a severe threat to human health worldwide. Early etiological diagnosis plays a critical role in controlling COVID-19 pandemic. However, etiological diagnosis has been largely compromised by high "false-negative" rates of viral nucleic acid testing, resulting from limited sampling efficiency using conventional oropharyngeal swabs. Here, we engineer regular swabs by using a microneedle (MN) patch to significantly improve the quality and quantity of virus collection. The combination of MNs with different crosslinking levels endows the patches with dual capability of mucus penetration and virus extraction. Moreover, the antibody (Ab) against viral spike protein was integrated into the patch, conferring MNs with an active virus capture potential. By taking advantage of the biological and engineered species, we believe that the designed MN/Ab swabs could serve as a promising tool to improve current sampling efficiency with fewer false negatives, contributing to the containment of the COVID-19 pandemic.
| S-EPMC7535620 | biostudies-literature
Project description:Many proteins work together with others in groups called complexes in order to achieve a specific function. Discovering protein complexes is important for understanding biological processes and predict protein functions in living organisms. Large-scale and throughput techniques have made possible to compile protein-protein interaction networks (PPI networks), which have been used in several computational approaches for detecting protein complexes. Those predictions might guide future biologic experimental research. Some approaches are topology-based, where highly connected proteins are predicted to be complexes; some propose different clustering algorithms using partitioning, overlaps among clusters for networks modeled with unweighted or weighted graphs; and others use density of clusters and information based on protein functionality. However, some schemes still require much processing time or the quality of their results can be improved. Furthermore, most of the results obtained with computational tools are not accompanied by an analysis of false positives. We propose an effective and efficient mining algorithm for discovering highly connected subgraphs, which is our base for defining protein complexes. Our representation is based on transforming the PPI network into a directed acyclic graph that reduces the number of represented edges and the search space for discovering subgraphs. Our approach considers weighted and unweighted PPI networks. We compare our best alternative using PPI networks from Saccharomyces cerevisiae (yeast) and Homo sapiens (human) with state-of-the-art approaches in terms of clustering, biological metrics and execution times, as well as three gold standards for yeast and two for human. Furthermore, we analyze false positive predicted complexes searching the PDBe (Protein Data Bank in Europe) database in order to identify matching protein complexes that have been purified and structurally characterized. Our analysis shows that more than 50 yeast protein complexes and more than 300 human protein complexes found to be false positives according to our prediction method, i.e., not described in the gold standard complex databases, in fact contain protein complexes that have been characterized structurally and documented in PDBe. We also found that some of these protein complexes have recently been classified as part of a Periodic Table of Protein Complexes. The latest version of our software is publicly available at http://doi.org/10.6084/m9.figshare.5297314.v1.
| S-EPMC5609739 | biostudies-literature
Project description:Objectives: With the worldwide spread of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), various antibody detection kits have been developed to test for SARS-CoV-2- specific IgG, IgM, and total antibody. However, the use of different testing methods under various heat-inactivation conditions might affect the COVID-19 detection results. Methods: Seven different antibody detection kits produced by four manufacturers for detection of SARS-CoV-2 IgG, IgM, and total antibody were tested at Wuhan Huoshenshan Hospital, China. Most of the kits used the indirect immunity, capture, and double-antigen sandwich methods. The effects of various heat-inactivation conditions on SARS-CoV-2-specific IgG, IgM, and total antibody detection were analyzed for the different test methods. Results: Using the indirect immunity method, values for SARS-CoV-2 IgG antibody significantly increased and those for IgM antibody decreased with increasing temperature of heat-inactivation using indirect immunity method. However, values for SARS-CoV-2 IgM and total antibody showed no change when the capture and double-antigen sandwich methods were used. The changes in IgG and IgM antibody values with the indirect immunity method indicated that heat-inactivation could affect COVID-19 detection results obtained using this method. In particular, 18 (22.2%) SARS-CoV-2 IgM positive samples were detected as negative with heat-inactivation at 65°C for 30 min, and one (25%) IgG negative sample was detected as positive after heat-inactivation at 56°C for 60 min and 60°C for 30 min. Conclusions: Heat-inactivation could increase SARS-CoV-2 IgG antibody values, and decrease IgM antibody values, causing potential false-positive or false-negative results for COVID-19 antibody detection using the indirect immunity method. Thus, before conducting antibody testing, the testing platforms should be evaluated in accordance with the relevant requirements to ensure accurate COVID-19 detection results.
| S-EPMC7849051 | biostudies-literature
Project description:BackgroundThe strength of selective constraints operating on amino acid sites of proteins has a multifactorial nature. In fact, amino acid sites within proteins coevolve due to their functional and/or structural relationships. Different methods have been developed that attempt to account for the evolutionary dependencies between amino acid sites. Researchers have invested a significant effort to increase the sensitivity of such methods. However, the difficulty in disentangling functional co-dependencies from historical covariation has fuelled the scepticism over their power to detect biologically meaningful results. In addition, the biological parameters connecting linear sequence evolution to structure evolution remain elusive. For these reasons, most of the evolutionary studies aimed at identifying functional dependencies among protein domains have focused on the structural properties of proteins rather than on the information extracted from linear multiple sequence alignments (MSA). Non-parametric methods to detect coevolution have been reported to be especially susceptible to produce false positive results based on the properties of MSAs. However, no formal statistical analysis has been performed to definitively test the differential effects of these properties on the sensitivity of such methods.ResultsHere we test the effect that variations on the MSA properties have over the sensitivity of non-parametric methods to detect coevolution. We test the effect that the size of the MSA (number of sequences), mean pairwise amino acid distance per site and the strength of the coevolution signal have on the ability of non-parametric methods to detect coevolution. Our results indicate that all three factors have significant effects on the accuracy of non-parametric methods. Further, introducing statistical filters improves the sensitivity and increases the statistical power of the methods to detect functional coevolution. Statistical analysis of the physico-chemical properties of amino acid sites in the context of the protein structure reveals striking dependencies among amino acid sites. Results indicate a covariation trend in the hydrophobicities and molecular weight characteristics of amino acid sites when analysing a non-redundant set of 8000 protein structures. Using this biological information as filter in coevolutionary analyses minimises the false positive rate of these methods. Application of these filters to three different proteins with known functional domains supports the importance of using biological filters to detect coevolution.ConclusionCoevolutionary analyses using non-parametric methods have proved difficult and highly prone to provide spurious results depending on the properties of MSAs and on the strength of coevolution between amino acid sites. The application of statistical filters to the number of pairs detected as coevolving reduces significantly the number of artifactual results. Analysis of the physico-chemical properties of amino acid sites in the protein structure context reveals their structure-dependent covariation. The application of this known biological information to the analysis of covariation greatly enhances the functional coevolutionary signal and removes historical covariation. Simultaneous use of statistical and biological data is instrumental in the detection of functional amino acid sites dependencies and compensatory changes at the protein level.
| S-EPMC2362121 | biostudies-literature
Project description:ObjectiveAiming to improve the feasibility and reliability of using high-frequency oscillations (HFOs) for translational studies of epilepsy, we present a pipeline with features specifically designed to reject false positives for HFOs to improve the automatic HFO detector.MethodsWe presented an integrated, multi-layered procedure capable of automatically rejecting HFOs from a variety of common false positives, such as motion, background signals, and sharp transients. This method utilizes a time-frequency contour approach that embeds three different layers including peak constraints, power thresholds, and morphological identification to discard false positives. Four experts were involved in rating detected HFO events that were randomly selected from different posttraumatic epilepsy (PTE) animals for a comprehensive evaluation.ResultsThe algorithm was run on 768-h recordings of intracranial electrodes in 48 PTE animals. A total of 453 917 HFOs were identified by initial HFO detection, of which 450 917 were implemented for HFO refinement and 203 531 events were retained. Random sampling was used to evaluate the performance of the detector. The HFO detection yielded an overall accuracy of 0.95±0.03 , with precision, recall, and F1 scores of 0.92±0.05 , 0.99±0.01 , and 0.94±0.03 , respectively. For the HFO classification, our algorithm obtained an accuracy of 0.97±0.02 . For the inter-rater reliability of algorithm evaluation, the agreement among four experts was 0.94±0.03 for HFO detection and 0.85±0.04 for HFO classification.SignificanceOur approach shows that a segregated pipeline design with a focus on false-positive rejection can improve the detection efficiency and provide reliable results. This pipeline does not require customization and uses fixed parameters, making it highly feasible and translatable for basic and clinical applications of epilepsy.
| S-EPMC9712470 | biostudies-literature
Project description:Glycosylation is an important post-translational modification characterized by extensive heterogeneity. While mass spectrometry (MS) is the premier technique to characterize glycoproteins, the study of intact glycopeptides presents challenges often not observed in the study of unmodified species. High-field asymmetric waveform ion mobility spectrometry (FAIMS) involves in-line gas-phase separation and offers the potential to analyze glycopeptides without prior enrichment. While FAIMS has been assessed for its use in glycoproteomics, most of these studies focus heavily or exclusively on N-glycoproteomics. Therefore, we evaluated FAIMS for O-glycoprotein and mucin analysis. We generated three samples: the mucin-domain glycoprotein podocalyxin, a recombinant glycoprotein mixture, and a WGA enriched human platelet sheddome. Although FAIMS allowed for the identification of new podocalyxin O-glycosites, it yielded less glycopeptide identifications and glycoforms. FAIMS seemed to be more useful in the increasingly complex samples, since we observed a higher number of identified glycosylated species. Because of the labile nature of glycosidic bonds, and presence of a buffer gas in FAIMS separation, we investigated the possibility of increased in-source fragmentation during FAIMS analysis. We compared total ion chromatograms of identifications eluting within a minute of identifications bearing larger glycan structures, albeit with the same peptide backbone. FAIMS experiments showed a 2-5-fold increase in spectral matches from in-FAIMS fragmentation compared to control experiments. These results were also replicated in other previously published data, indicating that this is a systematic behavior. In summary, our study highlights that, although there are potential benefits gained when using FAIMS separation, caution must be exercised when using FAIMS due to in-FAIMS fragmentation, which limits its applicability in the field of O-glycoproteomics.
2023-09-11 | PXD041217 | Pride