12023 research outputs found
Sort by
From Pure Cultures to Particles: Tracing Microbial Metabolism Through Amino Acid ²H/¹H Ratios
No full textMicrobial metabolisms exert profound impact on our planet’s atmosphere and surface geochemistry. Most available tools to study microbial metabolism in the environment provide only snapshots of activity at the time of sampling. However, holistic understanding of microbial function requires the ability to quantitatively reconstruct their activities prior to sampling, for which tools are currently limited. The overarching research presented in this thesis addresses this challenge through development of a new isotopic tool, amino acid hydrogen isotope (δ2HAA) analysis, into a useful tracer of microbial metabolism in the environment. We begin by solving a major analytical challenge: correcting for contributions of exchangeable amine-bound hydrogen in derivatized amino acids, which unlocks the ability to accurately measure δ2HAA values in organisms via gas chromatography-pyrolysis-isotope ratio mass spectrometry. We demonstrate in aerobic heterotrophic bacteria and phytoplankton that δ2HAA values are controlled by metabolism (specifically, carbon flow in cells), and we apply this isotopic tool to natural samples of marine particulate organic matter (POM), demonstrating substantial potential turnover of photoautotrophic proteins into heterotrophic proteins (up to 57 ± 18%) in POM with depth at different ocean sites. We further explore the microscale dynamics of marine bacteria on diatom aggregates to contextualize our understanding of controls on marine POM degradation. In particular, we find that both intra- and interspecies interactions profoundly shape microbial colonization dynamics, which in turn likely affect bulk particle degradation rates. Together, this body of work demonstrates the profound utility of δ2HAA analysis as a tracer of microbial metabolism—a timely development given the need to trace and quantify the metabolic responses of microbial communities to ongoing environmental perturbations
Signal Processing for Line Spectra: New Sensor Arrays, Algorithms, and Theoretical Results
Get PDFLine spectrum signals appear in diverse application areas such as molecular dynamics, power electronics, speech processing, and target localization. They are composed of sums of complex exponentials with distinct frequencies. Identifying the parameters of these constituent complex exponentials has been a prominent research topic in signal processing for over four decades. In this thesis, we focus on two specific applications involving line spectrum signals: direction of arrival (DOA) estimation using sensor arrays, and denoising of discrete-time periodic signals.
The main contribution of this thesis on the topic of DOA estimation is to propose unconventional sensor array geometries and algorithms in the presence of aperture constraints. In the first part, we demonstrate that under an aperture constraint, the traditional integer arrays (defined as arrays with sensors placed at integer multiples of the half-wavelength distance λ/2) can perform only suboptimally because of the restrictive sensor placement at integer locations. To address this, we propose to use 'rational arrays' that can have sensors located at rational multiples of λ/2. This offers greater flexibility in sensor placement under aperture constraints. In particular, we propose rational coprime arrays that can approach the Cramér-Rao bound (CRB) even at low signal-to-noise ratio (SNR) and with a limited number of snapshots, and can outperform the integer arrays. Numerical simulations show that rational arrays are also better equipped to resolve closely separated DOAs. To enable the derivation of the theoretical results and identifiability guarantees for rational coprime arrays, we extend the number-theoretic concepts such as greatest common divisor and coprimality to the case of rational numbers, and prove several number-theoretic properties. Rational arrays are also demonstrated to have important advantages when the DOAs are known to lie in a sector of the space, and for identifying O(N2) uncorrelated sources using N sensors under aperture constraint.
In the second part of the thesis, we propose modifications to the traditionally used sparse (integer) array design criteria. These modifications are aimed at mitigating the impact of mutual coupling on DOA estimation and reducing the required aperture. To reduce the impact of mutual coupling, we propose two types of sparse arrays that have either double or triple minimum inter-element spacing compared to the traditionally used λ/2 spacing. This introduces 'holes' at lags 1 and 2 in the difference coarrays (defined as the set of differences in sensor locations). The first type of arrays, called weight-constrained sparse arrays, have O(N) aperture, making them suitable when the available aperture is constrained and the number of DOAs is small. A general array construction, to further reduce the weights at other coarray lags, is also proposed. The second type of arrays, called weight-constrained nested arrays, have O(N2) degrees of freedom and are suitable when there are no aperture restrictions. Extensive Monte-Carlo simulations demonstrate that the proposed arrays have significantly smaller DOA estimation errors compared to the well-known sparse arrays from the literature, in the presence of high mutual coupling.
Because of the central holes in the difference coarrays of the weight-constrained arrays, there are two segments of consecutive entries in their coarrays: one on the positive side and the other on the negative side. To leverage these both, we propose to use an augmented coarray covariance matrix for the subspace-based algorithms such as multiple signal classification (MUSIC) and estimation of signal parameters via rotational invariance (ESPRIT). This further reduces the DOA estimation error for the weight-constrained arrays, and the computation time of augmented-MUSIC is significantly less than that of optimization-based methods, such as coarray interpolation and dictionary-based methods. We also develop methods to algorithmically interpolate the missing entries in the coarray at lags 1 and 2, to generate a larger coarray matrix. This approach demonstrates the capability to identify up to twice as many DOAs compared to what can be achieved using only the one-sided segment of consecutive lags in the coarray. This mitigates the main disadvantage of having central holes in the coarrays of weight-constrained arrays, while still benefiting from their advantage in reducing the impact of mutual coupling.
One major drawback of using coarray-MUSIC for DOA estimation is its inefficiency (i.e., the mean squared error (MSE) does not approach CRB, even asymptotically). We conduct several experiments to provide new insights into the complex relationship of coarray-MUSIC MSE on several parameters, such as array geometry, DOA separation, and accuracy of the estimated array output correlations. Furthermore, we demonstrate that an alternative way of constructing the Toeplitz covariance matrix can greatly improve the MSE compared to coarray-MUSIC, and can lead to efficient DOA estimation. This approach is based on solving an optimization problem whose objective is derived using the asymptotic error distribution of the known entries from the covariance matrix. We also propose a modification to the Toeplitz covariance matrix construction approach to account for the presence of mutual coupling and provide simulations with different sparse arrays.
The third part of the thesis is focused on developing a periodicity-aware signal denoising framework using Capon-optimized Ramanujan filter banks and pruned Ramanujan dictionaries. The signal reconstruction (synthesis) is done by solving a regularized optimization problem, based on the outputs of the analysis filter bank. This hybrid analysis-synthesis framework ensures that the denoised output is necessarily composed of discrete-time periodic components. Capon beamforming principles from array signal processing are utilized to optimize the Ramanujan filters to the incoming data. A computationally efficient way of obtaining the inverses of the required autocorrelation matrices is derived using Levinson’s recursion. The proposed denoising method is observed to be effective even when the signal length is small and demonstrates a high SNR gain across a wide range of input signal SNRs. Furthermore, we derive several decimation properties of Ramanujan subspace signals, which help in reducing the required computations by appropriately downsampling the filter outputs without any loss of information.
Towards the end of the thesis, we theoretically investigate the locations of zeros of Ramanujan filters. Additionally, we propose an ideal interpolation filter model for Ramanujan subspace signals, which has potential application in developing a synthesis filter bank counterpart to the Ramanujan analysis filter bank for perfect signal reconstruction. We also explore the use of dictionary learning to represent periodic signals, and adapt a convolutional neural network based DOA estimation method to sparse arrays.</p
Effects of Disorder on Quantum Phase Transitions and Quantum Dynamics
No full textWe present experimental studies of the effects of disorder on the quantum phase transitions of antiferromagnetic LiErF₄ and of the dynamic behavior of LiHo0.2Y0.8F₄, which hosts a spin glass ground state due to the combination of substitutional disorder and magnetic frustration. Both compounds are insulating dipolar-coupled magnets that can be effectively treated as spin ½ systems.
Two distinct quantum phase transitions can be induced in the easy-plane antiferromagnet LiErF₄, applying a magnetic field in the plane or perpendicular to it. The isotopic distribution of natural Er permits us to probe these transitions in the clean and dirty regimes. 167Er has a natural abundance of 23% and is the only stable isotope with a non-zero nuclear spin. At low temperatures, the nuclear spin slaves to the electronic spin and reduces the effective field felt by the electronic spin, thereby inducing random mass disorder in the dirty (low-temperature) regime. We use specific heat measurements to identify the temperature scale of the crossover between the dirty and clean regime as T=150 mK, and make ac magnetic susceptibility measurements to characterize the effects of disorder on the two quantum phase transitions. When the field is applied along the c-axis, the critical behavior is consistent with a violation of the Harris criterion in the clean regime and a change of universality class in the dirty regime. When the field is applied along the a-axis, the critical behavior is unchanged by the crossover between clean and dirty regimes.
We use ac susceptibility measurements to conduct thermal memory dip experiments on the spin glass state of LiHo0.2Y0.8F₄ in zero magnetic field and find no apparent rejuvenation or memory. We perform an analogous “quantum memory dip” measurement which uses a transverse magnetic field rather than temperature to enter the spin glass state, and we find strong rejuvenation. The relaxation rate of the susceptibility decreases as the transverse field increases. This counterintuitive result is attributed to an increase in the variance of the random longitudinal field associated with increasing the transverse field and is supported by simulations. Finally, we perform a "negative field cycle" experiment which finds erasure of memory in the spin glass state. We establish a theoretical framework of quantum resonant tunneling to explain our results, rather than the conventional picture of a hierarchical free energy landscape associated with classical spin glasses.</p
Methods for Long Read RNA-Seq Transcriptomics
No full textWhile short read RNA-seq dominated the field for decades, long read RNA-seq is particularly useful for isoform-level expression analysis, genome annotation, detecting novelly splicing transcripts, identifying exact breakpoints in gene fusions, and discovering chimeric RNAs. Long read RNA-seq has rapidly scaled to the point of producing terabytes of data from a single set of experiments. Technological advances in RNA and DNA sequencing library preparation, chemistry used in the Oxford nanopores, and basecalling algorithms have reduced long read sequencing error rates to sub-1% error. Further, the cost of long read sequencing has dropped to about one hundred US dollars per human genome. These two factors have lead to the mass production of high-throughput, long read, and single-cell RNA-seq data. While recent tools for long read RNA-seq have been developed, they have not kept pace in scalability and accuracy with long read RNA-seq in the fashion that short read RNA-seq tools have met computational scalability and accuracy challenges. To address this, in this thesis, we leverage long k-mers and pseudoalignment for mapping and quantifying long reads in the novel algorithm implemented within lr-kallisto, which yields both efficiency and higher accuracy for long read mapping and quantification than previous tools. We demonstrate that long read RNA-seq has reached sufficient depth and accuracy to yield accurate quantification of isoform-level expression for differential expression analysis. Furthermore, we explore the feasibilty of also utilizing long k-mers and pseudoalignment in both transcript discovery in dn-kallisto and gene fusion and immune receptor sequence discovery with fugi with measured success. Thus, our tools will enable a more complete, accurate, and scalable analysis of single-cell and bulk RNA-seq than has hitherto been possible in both quantifications and differential expression analysis as well as investigation of gene fusions, chimeric RNAs, and immune receptor sequences without bias
Domestication of Environmental Bacteria for Biosensing Applications
No full textThe field of synthetic biology has made impressive progress in the past 25 years, but is still lacking when it comes to our capability to predictably engineer organisms outside of a small group of lab model organisms. In this thesis, I present the efforts to domesticate two soil bacteria important in agriculture for biosensing. The first, Pseudomonas synxantha, a wheat-colonizing bacterium that helps fight off fungal disease, was engineered into a bioreporter for phosphorus limitation. We also made cell-free extract from this organism, to enable rapid characterization of genetic elements. For the second, Xenorhabdus griffiniae, we asked the question of whether this bacterium can sense the presence of its entomopathogenic nematode host Steinernema hermaphroditum. We learned that X. griffiniae is able to sense its host and we were able to build an early variant of a nematode reporter by first characterizing genetic elements in X. griffiniae
The Topology of Cellular Ontogeny
No full textA fundamental goal of modern biology is to build global, predictive models of gene regulation that encompass diverse physiological contexts. Single-cell transcriptomics has enabled the creation of developmental cell atlases--detailed catalogs of gene expression patterns and differentiation trajectories at an organismal scale. The widespread availability of cell atlases across metazoan model organisms presents an opportunity to construct global theories of cell-state control. In this thesis, we introduce a framework that uses persistent homology to decompose cell atlases into topological structures that provide signatures of gene regulation at the scale of an organism. Using this framework, we found that the topological structure of a broad set of developmental atlases contains only a discrete set of topological structures—such as clusters, trees, and loops—-revealing the recurrent use of global gene regulatory strategies. Our analysis revealed that the tree topology, while predominant, is not universal. Indeed, we identified non-trivial topologies containing loops in the development of human immune cells, seam-hypodermal cells in \textit{C. elegans}, and the cnidocytes of multiple cnidarians. Analysis of cell-state manifolds with non-trivial topology demonstrated an important role of convergent structures in increasing cellular diversity along paths to a common cell fate, and of cyclic structures in self-renewal of progenitor-like states. Together, this work provides a global perspective on principles of cell-state regulation, and suggests that loops are important organizing structures for controlling cell differentiation
Mixing-Driven Abyssal Ocean Circulation over Sloping Topography
No full textThe planetary-scale overturning circulation of the ocean is maintained by small-scale diapycnal mixing in the abyss. Recent theory and observations suggest that this turbulence is bottom-enhanced, confining the upwelling needed to close this circulation to thin bottom boundary layers (BLs) over sloping topography. Developing an understanding of how this mixing shapes the abyssal circulation, both locally and at the basin scale, is the unifying goal of this thesis.
The local response of a water column to mixing has previously been understood using a one-dimensional model of a rotating, stratified fluid over a sloping seafloor. Canonically, this model assumes no cross- or along-slope variations of the flow, pressure, and buoyancy anomalies. At steady state, it predicts a peculiar form of the net cross-slope transport, however, failing to consider its coupling to the global circulation. For symmetric bathymetry without along-slope variations, for instance, this large-scale context implies that all cross-slope BL transport must be exactly returned in the interior. This interior downwelling is then turned by the Coriolis acceleration, rapidly spinning up along-slope flow in balance with a cross-slope barotropic pressure gradient. With these added physics, the one-dimensional model better captures the local response to mixing over an idealized ridge, for example. Using BL theory, we explicitly describe how the BL and interior communicate in this model. The up-slope transport of dense water in the bottom BL contributes a net downward flux of buoyancy, creating an effective bottom boundary condition on the interior. The coupling goes both ways, with the interior stratification at the top of the BL setting the strength of the BL transport. Variations across the slope then allow for BL--interior exchange.
Ultimately, the net transport of the local response must conserve potential vorticity at the basin scale. To better understand this coupling for arbitrary topography, we develop a novel finite element model of the planetary geostrophic equations. Using a combination of simulations and BL theory, we then study the mixing-driven abyssal circulation in an idealized bowl-shaped basin. In the absence of wind forcing and the joint effect of baroclinicity and relief, the leading-order barotropic transport flows along f/H contours, where f is the Coriolis frequency and H is the depth. The local response to mixing is coupled to this barotropic circulation, simultaneously constrained by the barotropic circulation and forcing it via a bottom stress curl. For closed f/H contours, a strong along-contour barotropic circulation spins up, reminiscent of the local response described above. On the other hand, if these contours intersect the boundary, a case more typical in the real ocean, the barotropic transport is suppressed. This decouples the leading-order local response from the large-scale circulation and intensifies bottom BL upwelling. This work therefore suggests that the local abyssal stratification in the presence of bottom-enhanced mixing strongly depends on the large-scale context.</p
Observational and Computational Studies of Atmospheric Particle Formation
Get PDFAerosols are a ubiquitous component of the atmosphere, playing pivotal roles in air quality and climate. This thesis explores the way these particles come to be, and their roles in these atmospheric processes.
Aerosols form from a variety of anthropogenic and biogenic activities, processes which are very prominent in urban settings. In Los Angeles, the last decade of research has been dominated by the role of summertime secondary organic aerosol (SOA) in contributing to particulate matter (PM). Here, we make observations in the equinox seasons and in the winter and detail the formation of atmospheric aerosols in these seasons. Using aerosol mass spectrometry, we demonstrate that ammonium nitrate persists as one of the dominant secondary aerosol components despite dramatic reductions in nitrogen oxide (NOₓ) emissions. Further, we show that this ammonium nitrate is not measured by routine air quality measurements, biasing regulatory PM2.5 measurements. In the wintertime, similar techniques demonstrate that primary organic aerosol, as opposed to secondary, is an important component of the PM2.5, contrary to the prevailing narratives that SOA dominates the aerosol mass.
At global scales, the role of these aerosols in cloud formation and climate processes is of primary interest. While a variety of physicochemical properties of aerosols are important in the formation of cloud droplets, we focus here on the specific process of organic surface-partitioning. It has been suggested that in phase-separated aerosol, organic-rich surface layers can depress the surface tension of the particles, lowering their barrier to activate into cloud droplets. We assess this propensity for surface tension depression in two SOA systems, α-pinene and β-caryophyllene. Synergizing laboratory measurements, a thermodynamic model, and field data, it is shown that surface-active organics in these SOA systems can impact their hygroscopicity, though perhaps not sufficiently to warrant inclusion of these processes in global-scale models.</p
Spin Geometry and Quantum Diffusion
Get PDFThis thesis studies diffusion processes on spinor endomorphism algebras. The spinor and connection laplacian generated heat semigroups are shown to quantum dynamical semigroups, and after spectral truncation the existence of Evans-Hudson flows is established. The vacuum state expectation of the process is related to the spectral action principle in noncommutative geometry. Examples where the flow is proven to exist for untruncated laplacians are given. Convergence of finite dimensional approximations, through discretization and truncation, to spectral triples encoding Riemannian geometry and their statespaces as quantum metric spaces is also considered
Quantum Gibbs Sampling
Get PDFMarkov Chain Monte Carlo algorithms are indispensable in classical thermodynamic simulation, perhaps due to their mathematical simplicity, algorithmic efficiency, and physical origin. In particular, Glauber dynamics is a detailed-balanced continuous-time Markov chain that fixes the Gibbs distribution and also serves as a mathematically succinct model of classical thermalization. In this thesis, we proposed a quantum computation analog of Glauber dynamics that is exactly detailed balanced yet algorithmic efficient, inherits the locality of the target Hamiltonian, and resembles Davies'-like generators physically derived from a weak system-bath coupling. We hope our proposal will serve as a quantum algorithm for quantum thermodynamic simulation and a model of open system thermalization where a suitable construction has been lacking for noncommuting Hamiltonians