Search NASA⌕ Search

SEARCH · Search NASA

Results for “Statistical Sampling Sequence”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records

Does Collection Time Bias the Ecology of Cleanroom Air Samples?

Microbial monitoring of astromaterials collections has taken on increased importance with the return of biologically sensitive samples from the asteroids Ryugu and Bennu and the initiation of the Mars Sample Return Program. Terrestrial bacteria and fungi can alter the mineralogy and organic composition of our collections causing irreversible contamination of pristine samples and increasing the risk of false positives for life detection measurements. NASA has conducted routine microbial monitoring of its existing collections since 20181. Initial monitoring focused on surface samples collected with foam swabs. Although, airborne microbiology is often decoupled from surface microbiology in the built environment2 culture-based air sampling techniques like impactors were not compliant with existing contamination control requirements. Bringing organic rich media, gelatin or liquids into curation cleanrooms presents an unacceptable risk to pristine samples. In 2022 NASA purchased a materials complaint air sampler and began collecting air samples from the cleanrooms in addition to surface samples3. The new instrument uses an electret filter to collect samples that are suitable for cultivating organisms or for direct DNA sequencing. Preliminary DNA sequencing results appeared to indicate that longer sampling times biased the microbial community in favor of hearty, spore-forming bacteria3. We present the results of a study comparing overnight sampling (17 hours) to short (1 hour) sampling of unoccupied curation cleanrooms. The results will help us optimize our monitoring protocols and develop a more detailed inventory of the ecology of astromaterials curation cleanrooms. Methods: We analyzed 72 paired air samples from six different cleanrooms including the meteorite processing lab (ISO 7 equivalent, 16 samples), the lunar lab (ISO 6 equivalent, 10 samples), the stardust lab (ISO 5 equivalent 14 samples), the OSIRIS-REx lab (ISO 5 equivalent, 12 samples), the Hayabusa2 lab (ISO 5 equivalent, 14 samples), and the Genesis lab (ISO 4 equivalent, 6 samples). All the samples were collected with an InnovaPrep Bobcat air sampler operating at a sampling rate of 200 L/min. The sampler operates for 5 minutes out of every 20 minute period. Half of the samples were collected by filtering 3,000L (15 min. of active sampling) of air across an electret filter for one hour. The rest of the samples were collected by filtering approximately 51,000 L air across the filter overnight (~17 hours, 255 min. of active sampling). Cells were eluted from the filter using 6-7 ml of pressurized 0.15% tween 20 in PBS (phosphate buffered saline). This liquid was used to cultivate bacteria according to previously published methods1,4,5 and for DNA extraction and next generation sequencing. DNA was extracted with a Qiagen MagAttract PowerMicrobiome kit6. To identify bacteria and archaea, the 16S rRNA gene was amplified using Earth Microbiome primers for the V4 region 7. The amplified DNA was sequenced on an Illumina MiSeq using a V3 reagent kit. The resulting sequences were processed using DADA2 and QIIME2 as implemented on the EDGE bioinformatics platform8–10. Results: Only two of the 72 samples had no amplifiable DNA. Amplified DNA concentrations ranged from 2.67 – 0.272 ng/µl. The median concentration of amplified DNA for the 1 hour samples was 0.770 ± 0.368 ng/µl. The median concentration of amplified DNA for the overnight samples was 0.877 ± 0.434 ng/µl. On average the overnight samples had slightly more sequences (58,960 vs. 59,456) and ASV’s (amplicon sequence variants) (60 vs 64.5) than the one hour samples, but these differences are not statistically significant. The most abundant ASV in every sample mapped to the genus Cupravidus. ASV’s mapping to the genuses Bacillus, Schlegelella, Thermus, and Staphylococcus were also common. Discussion and Future Work: Alpha diversity statistics like Shannon Entropy and Faith Phylogenetic Diversity are used to describe the diversity of organisms in a single sample. If a longer sampling time was biasing the data, we would expect to see a change in these diversity statistics vs. sample time. However, we did not observe this in our data. The median Shannon entropy was slightly higher for the overnight samples (3.773 vs 3.611) as was the Faith Phylogenetic Diversity (4.042 vs 3.596), but both values were within a standard deviation of each other for the two sampling times (Fig. 1). It is unlikely, that the longer sampling time is introducing bias into our data. We do observe a significant decrease in diversity when comparing the air samples by lab. The Genesis lab (ISO 4 equivalent) has a lower median number of ASV’s (45.5) than the other labs (62). Median values for Shannon Entropy (3.717 vs. 3.430) and Faith Phylogenetic Diversity (3.796 vs. 3.548) are also lower for Genesis, but those values are with one standard deviation of each other for the different sampling times. This is consistent with previous culture-based results suggesting that the environment in cleanrooms tends to select for a core group of organisms capable of surviving under dry, low nutrient, conditions. The presence of the ASV’s mapping to Cupravidus and Thermus in our sequencing blanks and controls suggests that several of the most common organisms in our samples represent contaminants from the reagents used to perform the DNA extractions and sequencing. Further work is needed to identify these contaminants, remove them from our data and recalculate the diversity statistics. This is a systematic error. Therefore, we do not expect removing the sequencing contaminants to change our conclusions. Longer air sample collection times appear to result in slightly higher diversity and do not bias the results towards “hardy” bacteria like spore-formers. Based on these preliminary results we conclude that sampling at least 3,000 liters of air is sufficient to capture the microbial diversity of cleanrooms, and that air samples can also be collected overnight without negatively impacting diversity. These results allow us to be flexible when designing microbial monitoring plans so that they do not interfere with routine lab activity. References: 1. Regberg, A. B. et al. 49th Lunar and Planetary Science Conference (2018). 2. The United States Pharmacopeial Convention. USP General Chapter <1116> (2013). 3. Regberg, A. B., et al. 54th Lunar and Planetary Science Conference (2023). 4. Regberg, A. B. et al. 53rd Lunar and Planetary Science Conference ( 2022). 5. Davis, R. E.,et al. 50th Lunar and Planetary Science Conference (2019). 6. Qiagen. MagAttract® PowerMicrobiome® DNA/RNA EP Kit Handbook. (2018). 7. Walters, W. et al. mSystems 1, (2015). 8. Callahan, B. J. et al. Nat. Methods 13, 581–583 (2016). 9. Hall, M. & Beiko, R. G. Microbiome Analysis: Methods and Protocols113–129 (Springer, 2018). 10. Philipson, C. et al. Bio-Protoc. 7, e2622 (2017).

A. B. Regberg↗

The Herschel DUNES Open Time Key Programme

We will use the unique photometric capabilities provided by Herschel to perform a deep and systematic survey for faint, cold debris disks around nearby stars. Our sensitivity-limited Open Time Key Programme (OTKP) aims at finding and characterizing faint extrasolar analogues to the Edgeworth-Kuiper Belt (EKB) in an unbiased, statistically significant sample of nearby FGK main-sequence stars. Our target set spans a broad range of stellar ages (from 0.1 to 10 Gyr) and is volume-limited (distances < 20 pc). All stars with known extrasolar planets within this distance are included; additionally, some M- and A-type stars will be observed in collaboration with the Herschel DEBRIS OTKP, so that the entire sample covers a decade in stellar mass, from 0.2 to 2 solar masses. We will perform PACS and SPIRE photometric observations covering the wavelength range from 70 to 500 microns. The PACS observations at 100 microns have been designed to detect the stellar photospheres down to the confusion limit with a signal-to-noise ratio > 5. The observations in the other Herschel bands will allow us to characterize, model, and constrain the disks. As a result, it will be possible for us to reach fractional dust luminosities of a few times 10-7, close to the EKB level in the Solar System. This will provide an unprecedented lower limit to the fractional abundance of planetesimal systems and allow us to assess the presence of giant planets, which would play dynamical roles similar to those played by Jupiter and Neptune in the Solar System. The proposed observations will provide new and unique evidence for the presence of mature planetary systems in the solar neighbourhood and, in turn, will address the universality of planet/planetary system formation in disks around young stars.

Danchi, William C.↗

Star counts from the Hubble Space Telescope Snapshot Survey. I - Galactic models

We report a photometric study of stars from 450 fields at high Galactic latitudes that were observed in the Hubble Space Telescope Snapshot Survey to an average limiting apparent magnitude of V = 21.4. There are 166 fields that contain quasars selected in radio, X-ray, and color-excess surveys. This sample of 273 stars is free of selection bias with respect to the density of stars. To within the Poisson errors, the total counts and magnitude distribution of this unbiased sample are in agreement with the Bahcall-Soneira model (Bahcall, 1986). The angular distribution of the faint stars favors, at the 2 sigma level, a somewhat steeper disk luminosity function and a smaller spheroid main-sequence normalization than given by the model. The sample does not have enough statistical power to distinguish between the two-component Bahcall-Soneira model and the three-component model first proposed by Gilmore and Reid (1983), which contains a thick disk. The statistical power of the survey would increase about 15-fold if colors were obtained for the stars: the data probe the main sequences of the disk, thick disk, and spheroid. Models with and without a thick disk could then be distinguished at the 6 sigma level. The HST Snapshot Survey includes an additional 284 fields, 279 of which are centered on quasars that were selected by objective-prism surveys. These 279 fields are expected to show and do exhibit bias against bright stars, making them unsuitable for testing Galactic models.

Gould, A.↗

A fully redundant double difference algorithm for obtaining minimum variance estimates from GPS observations

In double differencing a regression system obtained from concurrent Global Positioning System (GPS) observation sequences, one either undersamples the system to avoid introducing colored measurement statistics, or one fully samples the system incurring the resulting non-diagonal covariance matrix for the differenced measurement errors. A suboptimal estimation result will be obtained in the undersampling case and will also be obtained in the fully sampled case unless the color noise statistics are taken into account. The latter approach requires a least squares weighting matrix derived from inversion of a non-diagonal covariance matrix for the differenced measurement errors instead of inversion of the customary diagonal one associated with white noise processes. Presented is the so-called fully redundant double differencing algorithm for generating a weighted double differenced regression system that yields equivalent estimation results, but features for certain cases a diagonal weighting matrix even though the differenced measurement error statistics are highly colored.

Melbourne, William G.↗

Evaluation of Correction Methods for NASA GeneLab Transcriptomic Datasets

Conducting space biology experiments aboard the International Space Station, particularly those utilizing complex model organisms like mice, is expensive and difficult due to limited crew availability, hardware, and space. As a result, sample numbers from these studies are low, reducing the statistical power of any one experiment. Aggregating spaceflight datasets serves as a method to increase sample numbers, allowing for novel insights through bioinformatic analysis of ‘omics data from merged datasets. However, aggregating datasets can introduce unwanted variation including 1) differences in sample handling, processing, and sequencing platforms between datasets (technical variation) as well as 2) differences in experimental design between datasets. In the present study, NASA GeneLab-hosted RNAseq datasets from mouse liver tissues were used to evaluate several statistical methods to correct for this unwanted variation through two approaches, reference-based and standard. The following correction algorithms were applied with (reference-based) and/or without (standard) considering Universal Mouse RNA Reference samples: ComBat and ComBat_seq from the SVA package, median polish, empirical Bayes, and ANOVA-based algorithms from the MBatch package, and negative binomial regression normalization in the DESeq2 package. For each approach, after the correction algorithm was applied, differential gene expression (DGE) analysis of flight and ground control samples was performed with the combined data. The robustness of each tool was evaluated using BatchQC to determine statistical differences between datasets before and after correction, Principal Component Analysis to evaluate global gene expression in samples before and after correction, and by comparing DGE analysis of individual datasets and combined datasets before and after correction. The results showed that the reference-based approach introduced several additional (and likely artificial) DEGs when compared with the respective standard approach. Of the methods tested, standard ComBat and DESeq2 were identified as the most robust correction methods for combining spaceflight mouse liver RNAseq datasets hosted on GeneLab.

GeneLab↗

Absolute ages from crater statistics: Using radiometric ages of Martian samples for determining the Martian cratering chronology

In the absence of dates derived from rock samples, impact crater frequencies are commonly used to date Martian surface units. All models for absolute dating rely on the lunar cratering chronology and on the validity of its extrapolation to Martian conditions. Starting from somewhat different lunar chronologies, rather different Martian cratering chronologies are found in the literature. Currently favored models are compared. The differences at old ages are significant, the differences at younger ages are considerable and give absolute ages for the same crater frequencies as different as a factor of 3. The total uncertainty could be much higher, though, since the ratio of lunar to Martian cratering rate which is of basic importance in the models is believed to be known no better than within a factor of 2. Thus, it is of crucial importance for understanding the the evolution of Mars and determining the sequence of events to establish an unambiguous Martian cratering chronology from crater statistics in combination with clean radiometric ages of returned Martian samples. For the dating goal, rocks should be as pristine as possible from a geologically simple area with a one-stage emplacement history of the local formation. A minimum of at least one highland site for old ages, two intermediate-aged sites, and one very young site is needed.

Neukum, G.↗

Combining RNA-SEQ Datasets from NASA GENELAB: An Evaluation of Correction Methods

Background: Conducting space biology experiments aboard the International Space Station, particularly those utilizing complex model organisms like mice, is expensive and difficult due to limited crew availability, hardware, and space. As a result, sample numbers from these studies are low, reducing the statistical power of any one experiment. Aggregating spaceflight datasets serves as a method to increase sample numbers, allowing for novel insights through bioinformatic analysis of ‘omics data from merged datasets. However, aggregating datasets can introduce unwanted variation including 1) differences in sample handling, processing, and sequencing platforms between datasets (technical variation) as well as 2) differences in experimental design between datasets. Methods: In the present study, NASA GeneLab-hosted RNAseq datasets from mouse liver tissues were used to evaluate several statistical methods to correct for this unwanted variation through two approaches, reference-based and standard. The following correction algorithms were applied with (reference-based) and/or without (standard) considering Universal Mouse RNA Reference samples: ComBat and ComBat_seq from the SVA package, the median polish, empirical Bayes, and ANOVA-based algorithms from the MBatch package, and negative binomial regression normalization in the DESeq2 package. For each approach, after the correction algorithm was applied, differential gene expression (DGE) analysis of flight and ground control samples was performed with the combined data. The robustness of each tool was evaluated using BatchQC to determine statistical differences between datasets before and after correction, Principal Component Analysis to evaluate global gene expression in samples before and after correction, and by comparing DGE analysis of individual datasets and combined datasets before and after correction. Results: The results showed that the reference-based approach introduced several additional (and likely artificial) differentially expressed genes when compared with the respective standard approach. Conclusions: Of the methods tested, standard ComBat_seq and DESeq2 were identified as the most robust correction methods for combining spaceflight mouse liver RNAseq datasets hosted on GeneLab.

Finsam Samson↗

Batch Effect Correction Methods for NASA GeneLab Transcriptomic Datasets

RNA sequencing (RNA-seq) data from space biology experiments promise to yield invaluable insights into the effects of spaceflight on terrestrial biology. However, sample numbers from each study are low due to limited crew availability, hardware, and space. To increase statistical power, spaceflight RNA-seq datasets from different missions are often aggregated together. However, this can introduce technical variation or "batch effects", often due to differences in sample handling, sample processing, and sequencing platforms. Several computational methods have been developed to correct for technical batch effects, thereby reducing their impact on true biological signals. In this study, we combined 7 mouse liver RNA-seq datasets from NASA GeneLab (part of the NASA Open Science Data Repository) to evaluate several common batch effect correction methods (ComBat and ComBat-seq from the sva R package, and Median Polish, Empirical Bayes, and ANOVA from the MBatch R package). We quantitatively evaluated the ability of these methods to correct for technical batch variables in space biology RNA-seq data using the following criteria: BatchQC, principal component analysis, dispersion separability criterion, log fold change correlation, and differential gene expression analysis. Each batch variable / correction method combination was then assessed using a custom scoring approach to identify the optimal correction method for the combined dataset, by geometrically probing the space of all allowable scoring functions to yield an aggregate volume-based scoring measure. Finally, we describe the way in which the GeneLab multi-study analysis and visualization portal will allow users to examine the presence or absence of batch effects using multiple metrics. If the user chooses to perform batch effect correction, the scoring approach described here can be implemented to identify the optimal correction method to use for their specific combined dataset prior to analysis.

Lauren M. Sanders↗

Evaluation of Correction Methods for NASA GeneLab Transcriptomic Datasets

Conducting space biology experiments aboard the International Space Station, particularly those utilizing complex model organisms like mice, is expensive and difficult due to limited crew availability, hardware, and space. As a result, sample numbers from these studies are low, reducing the statistical power of any one experiment. Aggregating spaceflight datasets serves as a method to increase sample numbers, allowing for novel insights through bioinformatic analysis of ‘omics data from merged datasets. However, aggregating datasets can introduce unwanted variation including 1) differences in sample handling, processing, and sequencing platforms between datasets (technical variation) as well as 2) differences in experimental design between datasets such as sex or age of the model organism used. In the present study, NASA GeneLab-hosted RNAseq datasets from rodent liver tissues were used to evaluate several statistical methods to correct for this unwanted variation through two approaches, reference-based and standard. The following correction algorithms were applied with (reference-based) and/or without (standard) considering Universal Mouse RNA Reference samples: ComBat and ComBat_seq from the SVA package, median polish, empirical Bayes, and ANOVA-based algorithms from the MBatch package, and negative binomial regression normalization in the DESeq2 package. For each approach, after the correction algorithm was applied, differential gene expression (DGE) analysis of flight and ground control samples was performed with the combined data. The robustness of each tool was evaluated using BatchQC, to determine statistical differences between datasets before and after correction, Principal Component Analysis, to evaluate global gene expression in samples before and after correction, and by comparing DGE analysis of individual datasets and combined datasets before and after correction. The results showed that the reference-based approach introduced several additional (and likely artificial) DEGs when compared with the standard approach. Thus, the most robust standard correction will be implemented in the GeneLab Visualization 2.0 platform when datasets are combined.

GeneLab, RNA-seq, Batch Correction↗

A coronagraphic search for brown dwarfs around nearby stars

Brown dwarf companions have been searched for around stars within 10 pc of the Sun using the Johns-Hopkins University Adaptive Optics Coronagraph (AOC), a stellar coronagraph with an image stabilizer. The AOC covers the field around the target star with a minimum search radius of 1 sec .5 and a field of view of 1 arcmin sq. We have reached an unprecedented dynamic range of Delta m = 13 in our search for faint companions at I band. Comparison of our survey with other brown dwarf searches shows that the AOC technique is unique in its dynamic range while at the same time just as sensitive to brown dwarfs as the recent brown dwarf surveys. The present survey covered 24 target stars selected from the Gliese catalog. A total of 94 stars were detected in 16 fields. The low-latitude fields are completely dominated by background star contamination. Kolmogorov-Smirnov tests were carried out for a sample restricted to high latitudes and a sample with small angular separations. The high-latitude sample (b greater than or equal to 44 deg) appears to show spatial concentration toward target stars. The small separation sample (Delta Theta less than 20 sec) shows weaker dependence on Galactic coordinates than field stars. These statistical tests suggest that both the high-latitude sample and the small separation sample can include a substantial fraction of true companions. However, the nature of these putative companions is mysterious. They are too faint to be white dwarfs and too blue for brown dwarfs. Ignoring the signif icance of the statistical tests, we can reconcile most of the detections with distant main-sequence stars or white dwarfs except for a candidate next to GL 475. Given the small size of our sample, we conclude that considerably more targets need to be surveyed before a firm conclusion on the possibility of a new class of companions can be made.

Nakajima, T.↗

Analysis of delay reducing and fuel saving sequencing and spacing algorithms for arrival traffic

The air traffic control subsystem that performs sequencing and spacing is discussed. The function of the sequencing and spacing algorithms is to automatically plan the most efficient landing order and to assign optimally spaced landing times to all arrivals. Several algorithms are described and their statistical performance is examined. Sequencing brings order to an arrival sequence for aircraft. First-come-first-served sequencing (FCFS) establishes a fair order, based on estimated times of arrival, and determines proper separations. Because of the randomness of the arriving traffic, gaps will remain in the sequence of aircraft. Delays are reduced by time-advancing the leading aircraft of each group while still preserving the FCFS order. Tightly spaced groups of aircraft remain with a mix of heavy and large aircraft. Spacing requirements differ for different types of aircraft trailing each other. Traffic is reordered slightly to take advantage of this spacing criterion, thus shortening the groups and reducing average delays. For heavy traffic, delays for different traffic samples vary widely, even when the same set of statistical parameters is used to produce each sample. This report supersedes NASA TM-102795 on the same subject. It includes a new method of time-advance as well as an efficient method of sequencing and spacing for two dependent runways.

Neuman, Frank↗

Implementation and Testing of Turbulence Models for the F18-HARV Simulation

This report presents three methods of implementing the Dryden power spectral density model for atmospheric turbulence. Included are the equations which define the three methods and computer source code written in Advanced Continuous Simulation Language to implement the equations. Time-history plots and sample statistics of simulated turbulence results from executing the code in a test program are also presented. Power spectral densities were computed for sample sequences of turbulence and are plotted for comparison with the Dryden spectra. The three model implementations were installed in a nonlinear six-degree-of-freedom simulation of the High Alpha Research Vehicle airplane. Aircraft simulation responses to turbulence generated with the three implementations are presented as plots.

Yeager, Jessie C.↗

An automated approach to the design of decision tree classifiers

The classification of large dimensional data sets arising from the merging of remote sensing data with more traditional forms of ancillary data is considered. Decision tree classification, a popular approach to the problem, is characterized by the property that samples are subjected to a sequence of decision rules before they are assigned to a unique class. An automated technique for effective decision tree design which relies only on apriori statistics is presented. This procedure utilizes a set of two dimensional canonical transforms and Bayes table look-up decision rules. An optimal design at each node is derived based on the associated decision table. A procedure for computing the global probability of correct classfication is also provided. An example is given in which class statistics obtained from an actual LANDSAT scene are used as input to the program. The resulting decision tree design has an associated probability of correct classification of .76 compared to the theoretically optimum .79 probability of correct classification associated with a full dimensional Bayes classifier. Recommendations for future research are included.

Argentiero, P.↗

The Herschel* PEP-HERMES Luminosity Function- I. Probing the Evolution of PACS Selected Galaxies to z approx. equal to 4

We exploit the deep and extended far-IR data sets (at 70, 100 and 160 μm) of the Herschel Guaranteed Time Observation (GTO) PACS Evolutionary Probe (PEP) Survey, in combination with the Herschel Multi-tiered Extragalactic Survey data at 250, 350 and 500 μm, to derive the evolution of the rest-frame 35-, 60-, 90- and total infrared (IR) luminosity functions (LFs) up to z ∼ 4.We detect very strong luminosity evolution for the total IR LF (LIR ∝ (1 + z)(sup 3.55 +/- 0.10) up to z ∼ 2, and ∝ (1 + z)(sup 1.62 +/- 0.51) at 2 less than z less than approximately 4) combined with a density evolution (∝ (1 + z)(sup −0.57 +/- 0.22) up to z ∼ 1 and ∝ (1 + z)(sup −3.92 +/- 0.34) at 1 less than z less than approximately 4). In agreement with previous findings, the IR luminosity density (ρIR) increases steeply to z ∼ 1, then flattens between z ∼ 1 and z ∼ 3 to decrease at z greater than approximately 3. Galaxies with different spectral energy distributions, masses and specific star formation rates (SFRs) evolve in very different ways and this large and deep statistical sample is the first one allowing us to separately study the different evolutionary behaviours of the individual IR populations contributing to ρIR. Galaxies occupying the well-established SFR-stellar mass main sequence (MS) are found to dominate both the total IR LF and ρIR at all redshifts, with the contribution from off-MS sources (≥0.6 dex above MS) being nearly constant (∼20 per cent of the total ρIR) and showing no significant signs of increase with increasing z over the whole 0.8 < z <2.2 range. Sources with mass in the range 10 ≤ log(M/solar mass) ≤ 11 are found to dominate the total IR LF, with more massive galaxies prevailing at the bright end of the high-z (greater than approximately 2) LF. A two-fold evolutionary scheme for IR galaxies is envisaged: on the one hand, a starburst-dominated phase in which the Super Massive Black Holes (SMBH) grows and is obscured by dust (possibly triggered by a major merging event), is followed by an AGN-dominated phase, then evolving towards a local elliptical. On the other hand, moderately star-forming galaxies containing a low-luminosity AGN have various properties suggesting they are good candidates for systems in a transition phase preceding the formation of steady spiral galaxies.

Herschel↗

The Containment Assurance Risk Framework of the Mars Sample Return Program

The Mars Sample Return campaign aims at bringing rock and atmospheric samples from Mars to Earth through a series of robotic missions. These missions would collect the samples being cached and deposited on Martian soil by the Perseverance rover, place them in a container, and launch them into Martian orbit for subsequent capture by an orbiter that would bring them back. Given there exists a non-zero probability that the samples contain biological material, precautions are being taken to design systems that would break the chain of contact between Mars and Earth. These include techniques such as sterilization of Martian particles, redundant containment vessels, and a robust reentry capsule capable of accurate landings without a parachute. Requirements exist that the probability of containment not assured of Martian-contaminated material into Earth’s biosphere be less than one in a million. To demonstrate compliance with this strict requirement, a statistical framework was developed to assess the likelihood of containment loss during each sample return phase and make a statement about the total combined mission probability of containment not assured. The work presented here describes this framework, which considers failure modes or fault conditions that can initiate failure sequences ultimately leading to containment not assured. Reliability estimates are generated from databases, design heritage, component specifications, or expert opinion in the form of probability density functions or point estimates and provided as inputs to the mathematical models that simulate the different failure sequences. The probabilistic outputs are then combined following the logic of several fault trees to compute the ultimate probability of containment not assured. Given the multidisciplinary nature of the problem and the different types of mathematical models used, the statistical tools needed for analysis are required to be computationally efficient. While standard Monte Carlo approaches are used for fast models, a multi-fidelity approach to rare event probabilities is proposed for expensive models. In this paradigm, inexpensive low-fidelity models are developed for computational acceleration purposes while the expensive high-fidelity model is kept in the loop to retain accuracy in the results. This work presents an example of end-to-end application of this framework highlighting the computational benefits of a multi-fidelity approach.

Giuseppe Cataldo↗

A magnitude-limited survey of Cepheid companions in the ultraviolet

Results of a magnitude-limited survey of classic Cepheids brighter than 8th mag carried out to search for hot main-sequence companions are presented. Spectra of 76 stars obtained with the IUE satellite in the 2000-3200-A region were compared with the spectra of nonvariable supergiants and also the single Cepheid Delta Cep to search for excess flux at 2500 A from possible companions. Photometric companions were found for 21 percent of the sample. When the Cepheids known to be binary from either orbital motion or spectra in the 1200-2000-A region are included, the percentage of companions rises to 29 percent. If a statistical correction from stars with orbital motion is included, 34 percent have companions. This percentage is compared with that found by Abt et al. (1990) for B2-B5 main-sequence stars. If only systems with periods longer than a year and separations not more than 30 arcsec are considered, only 18 percent of the B stars will become Cepheids with companions.

Evans, Nancy R.↗

Multiclass Bayes error estimation by a feature space sampling technique

A general Gaussian M-class N-feature classification problem is defined. An algorithm is developed that requires the class statistics as its only input and computes the minimum probability of error through use of a combined analytical and numerical integration over a sequence simplifying transformations of the feature space. The results are compared with those obtained by conventional techniques applied to a 2-class 4-feature discrimination problem with results previously reported and 4-class 4-feature multispectral scanner Landsat data classified by training and testing of the available data.

Mobasseri, B. G.↗

Discovery of Activities via Statistical Clustering of Fixation Patterns

Human behavior often consists of a series of distinct activities, each characterized by a unique signature of visual behavior. This is true even in a restricted domain, such as piloting an aircraft, where patterns of visual signatures might represent activities like communicating, navigating, and monitoring. We propose a novel analysis method for gaze-tracking data, to perform blind discovery of these activities based on their behavioral signatures. The method is in some respects similar to recurrence analysis, but here we compare not individual fixations, but groups of fixations aggregated over a fixed time interval. The duration of this interval is a parameter that we will refer to as τ. We assume that the environment has been divided into a set of N different areas-of-interest (AOIs). For a given interval of time of duration τ, we compute the proportion of time spent fixating each AOI, resulting in an N-dimensional vector. These proportions can be converted to counts by multiplying by τ divided by the average fixation duration (another parameter that we fix at 280 milliseconds). We compare different intervals by computing the chi-square statistic. The p-value associated with the statistic is the likelihood of observing the data under the hypothesis that the data in the two intervals were generated by a single process with a single set of probabilities governing the fixation of each AOI. We have investigated the method using a set of 10 synthetic "activities," that sample 4 AOIs. Four of these activities visit 3 of the 4 AOIs, with equal probability; as there are four different ways to leave-one- out, there are four such activities. Similarly, there are six different activities that leave-two-out. Sequences of simulated behavior were generated by running each activity for 40 seconds, in sequence, for a total of 6.7 minutes. The figure to the right shows the matrix of chi-square statistics, using a value of 2.8 seconds for τ, corresponding to 10 fixations. Low values (dark) indicate poor evidence for activity differences, while high values (bright) indicate strong evidence. The dark squares along the main diagonal each correspond to the forty second intervals in which the activity was held constant; the 4x4 block at the lower left corresponds to the four leave-one-out activities, while the 6x6 block in the upper right corresponds to the leave-two-out activities. (The anti-diagonal pattern of white squares indicates those activity pairs that share no AOIs.) The chi-square values can be binarized by choosing a particular significance level; we are interested in grouping bins that represent the same activity, effectively accepting the null hypothesis. Therefore, we may adopt a relatively lax criterion; for example, choosing a p-value of 0.2 means that two behaviors that have only a 1-in-5 chance of being produced by a single activity might nevertheless be clustered together. We have explored several methods to perform clustering on the data and solving for the activity probabilities. Greedy methods begin by selecting the time bin that is similar to the most (or least) other bins, and then forming a cluster from it and all other non-discriminable bins. These methods show mediocre performance, as they do not take into account temporal contiguity. Preliminary results indicate that methods that "grow" clusters in time from seed points perform better.

activity analysis↗