Outlier Sequences Research Articles

Developing and validating an adaptive multi-layer vehicle trajectory reconstruction method for outlier removal

Trajectory data is vital for traffic flow studies, and aerial photography-based methods are increasingly used to collect such data. However, these datasets often contain errors from various sources, which can be exacerbated by numerical derivative processes. Previous efforts have not fully addressed some of these issues such as consistency, varying length of outlier sequences, and unknown ground truth trends. Moreover, existing validation methods are often indirect and problematic. To address these limitations, we propose an adaptive multi-layer vehicle trajectory reconstruction method, which consists of three modules: the Initial Window Arrangement module to ensure the precise alignment between the reconstruction window and detected outlier fragments, maintaining internal consistency at boundary points; the Window Size Feasibility Test module to adaptively determine the window size according to varying length of outlier sequences, and XGBoost-based Ground Truth Estimation module to be combined with a least-square-based objective function to significantly improve reconstruction accuracy and more closely replicate the underlying trend. Additionally, we introduce the jerk-based reconstruction, which outperforms the acceleration-based reconstruction. To reliably assess and select the best ground truth estimation scheme and objective function, a novel synthetic dataset containing both the ground truth and realistic outlier fragments is proposed. Subsequently, a comparative evaluation of six different outlier removal methods was conducted using Zen Traffic Data. The validation results, utilizing both the synthetic dataset and Zen Traffic Data, demonstrate the exceptional performance of the proposed method across various evaluation perspectives. The good performance of the proposed outlier removal method is further demonstrated by comparing the IDM calibration results using the trajectories with and without outliers being removed. With just one parameter (jerk anomaly threshold), the parameter settings of our method are more objective and generalizable.

PhylteR: Efficient Identification of Outlier Sequences in Phylogenomic Datasets.

In phylogenomics, incongruences between gene trees, resulting from both artifactual and biological reasons, can decrease the signal-to-noise ratio and complicate species tree inference. The amount of data handled today in classical phylogenomic analyses precludes manual error detection and removal. However, a simple and efficient way to automate the identification of outliers from a collection of gene trees is still missing. Here, we present PhylteR, a method that allows rapid and accurate detection of outlier sequences in phylogenomic datasets, i.e. species from individual gene trees that do not follow the general trend. PhylteR relies on DISTATIS, an extension of multidimensional scaling to 3 dimensions to compare multiple distance matrices at once. In PhylteR, these distance matrices extracted from individual gene phylogenies represent evolutionary distances between species according to each gene. On simulated datasets, we show that PhylteR identifies outliers with more sensitivity and precision than a comparable existing method. We also show that PhylteR is not sensitive to ILS-induced incongruences, which is a desirable feature. On a biological dataset of 14,463 genes for 53 species previously assembled for Carnivora phylogenomics, we show (i) that PhylteR identifies as outliers sequences that can be considered as such by other means, and (ii) that the removal of these sequences improves the concordance between the gene trees and the species tree. Thanks to the generation of numerous graphical outputs, PhylteR also allows for the rapid and easy visual characterization of the dataset at hand, thus aiding in the precise identification of errors. PhylteR is distributed as an R package on CRAN and as containerized versions (docker and singularity).

Fish forewarning of comprehensive toxicity in water environment based on Bayesian sequential method

Environmental impact of pollutants can be analyzed effectively by acquiring fish behavioral signals in water with biological behavior sensors. However, a variety of factors, such as the complexity of biological organisms themselves, the device error and the environmental noise, may compromise the accuracy and timeliness of model predictions. The current methods lack prior knowledge about the fish behavioral signals corresponding to characteristic pollutants, and in the event of a pollutant invasion, the fish behavioral signals are poorly discriminated. Therefore, we propose a novel method based on Bayesian sequential, which utilizes multi-channel prior knowledge to calculate the outlier sequence based on wavelet feature followed by calculating the anomaly probability of observed values. Furthermore, the relationship between the anomaly probability and toxicity is analyzed in order to achieve forewarning effectively. At last, our algorithm for fish toxicity detection is verified by integrating the data on laboratory acceptance of characteristic pollutants. The results show that only one false positive occurred in the six experiments, the present algorithm is effective in suppressing false positives and negatives, which increases the reliability of toxicity detections, and thereby has certain applicability and universality in engineering applications.

Defining and describing treatment heterogeneity in new-onset idiopathic lower back and extremity pain through reconstruction of longitudinal care sequences

BACKGROUND CONTEXTDespite established guidelines, long-term management of surgically-treated low back pain (LBP) and lower extremity pain (LEP) remains heterogeneous. Understanding care heterogeneity could inform future approaches for standardization of practices. PURPOSETo describe treatment heterogeneity in surgically-managed LBP and LEP. STUDY DESIGN/SETTINGRetrospective study of a nationwide commercial database spanning inpatient and outpatient encounters for enrollees of eligible employer-supplied healthcare plans (2007–2016). PATIENT SAMPLEA population-based sample of opioid-naïve adult patients with newly-diagnosed LBP or LEP were identified. Inclusion required at least 12-months of pre-diagnosis and post-diagnosis continuous follow-up. EXPOSUREIncluded treatments/evaluations include conservative management (chiropractic manipulative therapy, physical therapy, epidural steroid injections), imaging (x-ray, MRI, CT), pharmaceuticals (opioids, benzodiazepines), and spine surgery (decompression, fusion). OUTCOME MEASURESPrimary outcomes-of-interest were 12-month net healthcare expenditures (inpatient and outpatient) and 12-month opioid usage. METHODSAnalyses include interrogation of care sequence heterogeneity and temporal trends in sequence-initiating services. Comparisons were conducted in the framework of sequence-specific treatment sequences, which reflect the personalized order of healthcare services pursued by each patient. Outlier sequences characterized by high opioid use and costs were identified from frequently observed surgical treatment sequences using Mahalanobis distance. RESULTSA total of 2,496,908 opioid-naïve adult patients with newly-diagnosed LBP or LEP were included (29,519 surgical). In the matched setting, increased care sequence heterogeneity was observed in surgical patients (0.51 vs. 0.12 previously-unused interventions/studies pursued per month). Early opioid and MRI use has decreased between 2008 and 2015 but is matched by increases in early benzodiazepine and x-ray use. Outlier sequences, characterized by increased opioid use and costs, were found in 5.8% of surgical patients. Use of imaging prior to conservative management was common in patients pursuing outlier sequences compared to non-outlier sequences (96.5% vs. 63.8%, p<.001). Non-outlier sequences were more frequently characterized by early conservative interventions (31.9% vs. 7.4%, p<.001). CONCLUSIONSSurgically-managed LBP and LEP care sequences demonstrate high heterogeneity despite established practice guidelines. Outlier sequences associated with high opioid usage and costs can be identified and are characterized by increased early imaging and decreased early conservative management. Elements that may portend suboptimal longitudinal management could provide opportunities for standardization of patient care.

Estimation and prediction based on record statistics in the presence of an outlier

A single outlier sequence in which the distribution of the first observation differs from the others is considered and the properties of record statistics extracted from such sequence are studied. The problem of estimating the model parameters is discussed in the proportional hazard rate model. The maximum likelihood estimator and the uniformly minimum variance unbiased estimator are obtained for the special case of exponential distribution. The best linear unbiased (invariant) estimator is also derived for the location-scale family of distributions and their efficiencies are calculated. The problems of predicting the future records and reconstructing the past records are investigated. Various predictors and reconstructors are presented and some of their properties are stated. The precision of the obtained predictors are compared based on both the mean squared prediction error and Pitman’s measure of closeness criteria. Finally, an example with real data is given to illustrate the results.

Meteorlogical outliers detection based on artificial intelligence

To precisely evaluate the effect of artificial precipitation of Project Tianshui, anomaly data within the large dataset collected is supposed to be detected and dealt with reasonably, to enhance the analysis and prediction of the data. Using the data accumulated from 2008 to 2017 from Wushaoling Meteorological Station, and the data from May 2019 to October 2019 from the dedicated supervising network built for the project, taking temperature(0.1°C) and other meteorological data as example, anomaly dectection is conducted by the arithmetics Local Outlier Factor and Isolation Forest respectively. Consistency check is also conducted via the two arithmetics. The main outliers of the accumulated data for ten years are isolated outliers. LOF works pretty well, while Isolation Forest performs unsatisfying. The main outliers of the dedicated supervising network data are of a continuous outlier sequence. LOF can hardly detect the anomaly, while Isolation Forest precisely find out the outliers. Both arithmetics perform well on consistency check in space. The differences of LOF and Isolation Forest are compared. New ideas of using artificial intelligence arithmetics to detect meteorological outliers and pre-process the data are provided.

Genetic characterization of mumps viruses associated with the resurgence of mumps in the United States: 2015–2017

Despite high coverage with measles, mumps, and rubella vaccine in the United States, outbreaks of mumps occur in close contact settings such as schools, colleges, and camps. Starting in late 2015, outbreaks were reported from several universities, and by the end of 2017, greater than 13,800 cases had been reported nation-wide. In 2013, the CDC and the Association of Public Health Laboratories contracted four Vaccine Preventable Diseases Reference Centers (VPD-RCs) to perform real-time reverse transcription PCR (RT-qPCR) to detect mumps RNA in clinical samples and to determine the genotype. Twelve genotypes of mumps virus are currently recognized by the World Health Organization, and the standard protocol for genotyping requires sequencing the entire gene coding for the small hydrophobic (SH) protein. Phylogenetic analysis of the 1862 mumps samples genotyped from 2015 through 2017 showed that the overall diversity of genotypes detected was low. Only 0.8 % of the sequences were identified as genotypes C, H, J, or K, and 0.5 % were identified as vaccine strains in genotypes A or N, while most sequences (98.7 %) were genotype G. The majority of the genotype G sequences could be included into one of two large groups with identical SH sequences. Within genotype G, a small number of phylogenetically significant outlier sequences were associated with epidemiologically distinct chains of transmission. These results demonstrate that molecular and epidemiologic data can be used to track transmission pathways of mumps virus; however, the limited diversity of the SH sequences may be insufficient for resolving transmission in all outbreaks.

System log clustering approaches for cyber security applications: A survey

Log files give insight into the state of a computer system and enable the detection of anomalous events relevant to cyber security. However, automatically analyzing log data is difficult since it contains massive amounts of unstructured and diverse messages collected from heterogeneous sources. Therefore, several approaches that condense or summarize log data by means of clustering techniques have been proposed. Picking the right approach for a particular application domain is, however, non-trivial, since algorithms are designed towards specific objectives and requirements. This paper therefore surveys existing approaches. It thereby groups approaches by their clustering techniques, reviews their applicability and limitations, discusses trends and identifies gaps. The survey reveals that approaches usually pursue one or more of four major objectives: overview and filtering, parsing and signature extraction, static outlier detection, and sequences and dynamic anomaly detection. Finally, this paper also outlines a concept and tool that support the selection of appropriate approaches based on user-defined requirements.

Genomic sites hypersensitive to ultraviolet radiation

If the genome contains outlier sequences extraordinarily sensitive to environmental agents, these would be sentinels for monitoring personal carcinogen exposure and might drive direct changes in cell physiology rather than acting through rare mutations. New methods, adductSeq and freqSeq, provided statistical resolution to quantify rare lesions at single-base resolution across the genome. Primary human melanocytes, but not fibroblasts, carried spontaneous apurinic sites and TG sequence lesions more frequent than ultraviolet (UV)-induced cyclobutane pyrimidine dimers (CPDs). UV exposure revealed hyperhotspots acquiring CPDs up to 170-fold more frequently than the genomic average; these sites were more prevalent in melanocytes. Hyperhotspots were disproportionately located near genes, particularly for RNA-binding proteins, with the most-recurrent hyperhotspots at a fixed position within 2 motifs. One motif occurs at ETS family transcription factor binding sites, known to be UV targets and now shown to be among the most sensitive in the genome, and at sites of mTOR/5' terminal oligopyrimidine-tract translation regulation. The second occurs at A2-15TTCTY, which developed "dark CPDs" long after UV exposure, repaired CPDs slowly, and had accumulated CPDs prior to the experiment. Motif locations active as hyperhotspots differed between cell types. Melanocyte CPD hyperhotspots aligned precisely with recurrent UV signature mutations in individual gene promoters of melanomas and with known cancer drivers. At sunburn levels of UV exposure, every cell would have a hyperhotspot CPD in each of the ∼20 targeted cell pathways, letting hyperhotspots act as epigenetic marks that create phenome instability; high prevalence favors cooccurring mutations, which would allow tumor evolution to use weak drivers.

Sequence selection by FitSS4ASR alleviates ancestral sequence reconstruction as exemplified for geranylgeranylglyceryl phosphate synthase.

For evolutionary studies, but also for protein engineering, ancestral sequence reconstruction (ASR) has become an indispensable tool. The first step of every ASR protocol is the preparation of a representative sequence set containing at most a few hundred recent homologs whose composition determines decisively the outcome of a reconstruction. A common approach for sequence selection consists of several rounds of manual recompilation that is driven by embedded phylogenetic analyses of the varied sequence sets. For ASR of a geranylgeranylglyceryl phosphate synthase, we additionally utilized FitSS4ASR, which replaces this time-consuming protocol with an efficient and more rational approach. FitSS4ASR applies orthogonal filters to a set of homologs to eliminate outlier sequences and those bearing only a weak phylogenetic signal. To demonstrate the usefulness of FitSS4ASR, we determined experimentally the oligomerization state of eight predecessors, which is a delicate and taxon-specific property. Corresponding ancestors deduced in a manual approach and by means of FitSS4ASR had the same dimeric or hexameric conformation; this concordance testifies to the efficiency of FitSS4ASR for sequence selection. FitSS4ASR-based results of two other ASR experiments were added to the Supporting Information. Program and documentation are available at https://gitlab.bioinf.ur.de/hek61586/FitSS4ASR.

Global pentapeptide statistics are far away from expected distributions

The relationships between polypeptide composition, sequence, structure and function have been puzzling biologists ever since first protein sequences were determined. Here, we study the statistics of occurrence of all possible pentapeptide sequences in known proteins. To compensate for the non-uniform distribution of individual amino acid residues in protein sequences, we investigate separately all possible permutations of every given amino acid composition. For the majority of permutation groups we find that pentapeptide occurrences deviate strongly from the expected binomial distributions, and that the observed distributions are also characterized by high numbers of outlier sequences. An analysis of identified outliers shows they often contain known motifs and rare amino acids, suggesting that they represent important functional elements. We further compare the pentapeptide composition of regions known to correspond to protein domains with that of non-domain regions. We find that a substantial number of pentapeptides is clearly strongly favored in protein domains. Finally, we show that over-represented pentapeptides are significantly related to known functional motifs and to predicted ancient structural peptides.

Of daemons and men: reducing false positive rate in intrusion detection systems with file system footprint analysis

In this work, we propose a methodology for reducing false alarms in file system intrusion detection systems, by taking into account the daemon’s file system footprint. More specifically, we experimentally show that sequences of outliers can serve as a distinguishing characteristic between true and false positives, and we show how analysing sequences of outliers can lead to lower false positive rates, while maintaining high detection rates. Based on this analysis, we developed an anomaly detection filter that learns outlier sequences using k-nearest neighbours with normalised longest common subsequence. Outlier sequences are then used as a filter to reduce false positives on the $$FI^2DS$$ file system intrusion detection system. This filter is evaluated on both overlapping and non-overlapping sequences of outliers. In both cases, experiments performed on three real-world web servers and a honeynet show that our approach achieves significant false positive reduction rates (up to 50 times), without any degradation of the corresponding true positive detection rates.

Combining six genome scan methods to detect candidate genes to salinity in the Mediterranean striped red mullet (Mullus surmuletus)

BackgroundAdaptive genomics may help predicting how a species will respond to future environmental changes. Genomic signatures of local adaptation in marine organisms are often driven by environmental selective agents impacting the physiology of organisms. With one of the highest salinity level, the Mediterranean Sea provides an excellent model to investigate adaptive genomic divergence underlying salinity adaptation. In the present study, we combined six genome scan methods to detect potential genomic signal of selection in the striped red mullet (Mullus surmuletus) populations distributed across a wide salinity gradient. We then blasted these outlier sequences on published fish genomic resources in order to identify relevant potential candidate genes for salinity adaptation in this species.ResultsAltogether, the six genome scan methods found 173 outliers out of 1153 SNPs. Using a blast approach, we discovered four candidate SNPs belonging to three genes potentially implicated in adaptation of M. surmuletus to salinity. The allele frequency at one of these SNPs significantly increases with salinity independently from the effect of longitude. The gene associated to this SNP, SOCS2, encodes for an inhibitor of cytokine and has previously been shown to be expressed under osmotic pressure in other marine organisms. Additionally, our results showed that genome scan methods not correcting for spatial structure can still be an efficient strategy to detect potential footprints of selection, when the spatial and environmental variation are confounded, and then, correcting for spatial structure in a second step represents a conservative method.ConclusionThe present outcomes bring evidences of potential genomic footprint of selection, which suggest an adaptive response of M. surmuletus to salinity conditions in the Mediterranean Sea. Additional genomic data such as sequencing of a full-genome and transcriptome analyses of gene expression would provide new insights regarding the possibility that some striped red mullet populations are locally adapted to their saline environment.

Outlier detection and sequence reconstruction in continuous time series of ocean observation data based on difference analysis and the Dixon criterion

AbstractIn light of the specific characteristics of the long‐term record and complex sequence nature of the ocean observation data, a new method was developed based on the original Dixon detection criteria to be specifically detect and remove data outliers. This method combines the two traditional methods of data quality control and Dixon detection theory and assumes that the second‐order differential sequence of parameter measurements passes an appropriate stationarity test. Thus, the measurement attributes are considered to be in the same physical state and to occupy a small range of time and space, equivalent to a parallel observation test. Provided that the observations over a small range of time and space correspond to the record of a sequence covering a short period of time, this short time sequence is treated as a sliding window in the proposed new method. Outliers are detected based on lookup‐table after an index parameter Q is calculated within the sliding window. A correlation analysis and the test results show that the proposed new method can effectively instantiate a sequence of outliers characterized by different phases. Compared with other existing methods, the new method proved to be computationally efficient and easily programmable for practical implementation. Further, this method preserves the original data because the outliers are replaced by an inverse distance‐weighted average of the recorded data within the window, while other data were intact.

Universal sequential outlier hypothesis testing

ABSTRACTUniversal outlier hypothesis testing is studied in a sequential setting. Multiple observation sequences are collected, a small subset of which are outliers. A sequence is considered an outlier if the observations in that sequence are generated by an “outlier” distribution, distinct from a common “typical” distribution governing the majority of the sequences. Apart from being distinct, the outlier and typical distributions can be arbitrarily close. The goal is to design a universal test to best discern all the outlier sequences. A universal test with the flavor of the repeated significance test is proposed and its asymptotic performance, as the error probability goes to zero, is characterized under various universal settings. The proposed test is shown to be universally consistent. For the model with at most one outlier, conditioned on the outlier being present, the test is shown to be asymptotically optimal universally when the typical distribution is known and as the number of sequences goes to infinity when neither the outlier nor the typical distribution is known. With multiple identical outliers, the test is shown to be asymptotically optimal universally when the number of outliers is the largest possible and with the typical distribution being known, and its asymptotic performance with neither the outlier nor the typical distribution being known is also characterized. Extensions of the findings to models with multiple distinct outliers are also discussed. In all cases, it is shown that the asymptotic performance guarantees for the proposed test when neither the outlier nor the typical distribution is known converge to those when the typical distribution is known as the number of sequences goes to infinity.

Improving Functional Annotation in the DRE-TIM Metallolyase Superfamily through Identification of Active Site Fingerprints.

Within the DRE-TIM metallolyase superfamily, members of the Claisen-like condensation (CC-like) subgroup catalyze C-C bond-forming reactions between various α-ketoacids and acetyl-coenzyme A. These reactions are important in the metabolic pathways of many bacterial pathogens and serve as engineering scaffolds for the production of long-chain alcohol biofuels. To improve functional annotation and identify sequences that might use novel substrates in the CC-like subgroup, a combination of structural modeling and multiple-sequence alignments identified active site residues on the third, fourth, and fifth β-strands of the TIM-barrel catalytic domain that are differentially conserved within the substrate-diverse enzyme families. Using α-isopropylmalate synthase and citramalate synthase from Methanococcus jannaschii (MjIPMS and MjCMS), site-directed mutagenesis was used to test the role of each identified position in substrate selectivity. Kinetic data suggest that residues at the β3-5 and β4-7 positions play a significant role in the selection of α-ketoisovalerate over pyruvate in MjIPMS. However, complementary substitutions in MjCMS fail to alter substrate specificity, suggesting residues in these positions do not contribute to substrate selectivity in this enzyme. Analysis of the kinetic data with respect to a protein similarity network for the CC-like subgroup suggests that evolutionarily distinct forms of IPMS utilize residues at the β3-5 and β4-7 positions to affect substrate selectivity while the different versions of CMS use unique architectures. Importantly, mapping the identities of residues at the β3-5 and β4-7 positions onto the protein similarity network allows for rapid annotation of probable IPMS enzymes as well as several outlier sequences that may represent novel functions in the subgroup.

OD-seq: outlier detection in multiple sequence alignments.

BackgroundMultiple sequence alignments (MSA) are widely used in sequence analysis for a variety of tasks. Outlier sequences can make downstream analyses unreliable or make the alignments less accurate while they are being constructed. This paper describes a simple method for automatically detecting outliers and accompanying software called OD-seq. It is based on finding sequences whose average distance to the rest of the sequences in a dataset, is anomalous.ResultsThe software can take a MSA, distance matrix or set of unaligned sequences as input. Outlier sequences are found by examining the average distance of each sequence to the rest. Anomalous average distances are then found using the interquartile range of the distribution of average distances or by bootstrapping them. The complexity of any analysis of a distance matrix is normally at least O(N 2) for N sequences. This is prohibitive for large N but is reduced here by using the mBed algorithm from Clustal Omega. This reduces the complexity to O(N log(N)) which makes even very large alignments easy to analyse on a single core. We tested the ability of OD-seq to detect outliers using artificial test cases of sequences from Pfam families, seeded with sequences from other Pfam families. Using a MSA as input, OD-seq is able to detect outliers with very high sensitivity and specificity.ConclusionOD-seq is a practical and simple method to detect outliers in MSAs. It can also detect outliers in sets of unaligned sequences, but with reduced accuracy. For medium sized alignments, of a few thousand sequences, it can detect outliers in a few seconds. Software available as http://www.bioinf.ucd.ie/download/od-seq.tar.gz.

Universal Outlier Hypothesis Testing

Outlier hypothesis testing is studied in a universal setting. Multiple sequences of observations are collected, a small subset of which are outliers. A sequence is considered an outlier if the observations in that sequence are distributed according to an ``outlier'' distribution, distinct from the ``typical'' distribution governing the observations in all the other sequences. Nothing is known about the outlier and typical distributions except that they are distinct and have full supports. The goal is to design a universal test to best discern the outlier sequence(s). It is shown that the generalized likelihood test is universally exponentially consistent under various settings. The achievable error exponent is also characterized. In the other settings, it is also shown that there cannot exist any universally exponentially consistent test.

Beyond an AFLP genome scan towards the identification of immune genes involved in plague resistance in Rattus rattus from Madagascar

Genome scans using amplified fragment length polymorphism (AFLP) markers became popular in nonmodel species within the last 10 years, but few studies have tried to characterize the anonymous outliers identified. This study follows on from an AFLP genome scan in the black rat (Rattus rattus), the reservoir of plague (Yersinia pestis infection) in Madagascar. We successfully sequenced 17 of the 22 markers previously shown to be potentially affected by plague-mediated selection and associated with a plague resistance phenotype. Searching these sequences in the genome of the closely related species Rattus norvegicus assigned them to 14 genomic regions, revealing a random distribution of outliers in the genome (no clustering). We compared these results with those of an in silico AFLP study of the R. norvegicus genome, which showed that outlier sequences could not have been inferred by this method in R. rattus (only four of the 15 sequences were predicted). However, in silico analysis allowed the prediction of AFLP markers distribution and the estimation of homoplasy rates, confirming its potential utility for designing AFLP studies in nonmodel species. The 14 genomic regions surrounding AFLP outliers (less than 300 kb from the marker) contained 75 genes encoding proteins of known function, including nine involved in immune function and pathogen defence. We identified the two interleukin 1 genes (Il1a and Il1b) that share homology with an antigen of Y. pestis, as the best candidates for genes subject to plague-mediated natural selection. At least six other genes known to be involved in proinflammatory pathways may also be affected by plague-mediated selection.

Anomaly Detection and Diagnosis Algorithms for Discrete Symbol Sequences with Applications to Airline Safety

We present a set of novel algorithms which we call sequenceMiner that detect and characterize anomalies in large sets of high-dimensional symbol sequences that arise from recordings of switch sensors in the cockpits of commercial airliners. While the algorithms that we present are general and domain-independent, we focus on a specific problem that is critical to determining the system-wide health of a fleet of aircraft. The approach taken uses unsupervised clustering of sequences using the normalized length of the longest common subsequence as a similarity measure, followed by detailed outlier analysis to detect anomalies. In this method, an outlier sequence is defined as a sequence that is far away from the cluster center. We present new algorithms for outlier analysis that provide comprehensible indicators as to why a particular sequence is deemed to be an outlier. The algorithms provide a coherent description to an analyst of the anomalies in the sequence when compared to more normal sequences. In the final section of the paper, we demonstrate the effectiveness of sequenceMiner for anomaly detection on a real set of discrete-sequence data from a fleet of commercial airliners. We show that sequenceMiner discovers actionable and operationally significant safety events. We also compare our innovations with standard hidden Markov models, and show that our methods are superior.