Publications

The target of many astronomical studies is the recovery of tiny astrophysical signals living in a sea of uninteresting (but usually dominant) noise. In many contexts (i.e., stellar time-series, or high-contrast imaging, or stellar spectroscopy), there are structured components in this noise caused by systematic effects in the astronomical source, the atmosphere, the telescope, or the detector. More often than not, evaluation of the true physical model for these nuisances is computationally intractable and dependent on too many (unknown) parameters to allow rigorous probabilistic inference. Sometimes, housekeeping data---and often the science data themselves---can be used as predictors of the systematic noise. Linear combinations of simple functions of these predictors are often used as computationally tractable models that can capture the nuisances. These models can be used to fit and subtract systematics prior to investigation of the signals of interest, or they can be used in a simultaneous fit of the systematics and the signals. In this Note, we show that if a Gaussian prior is placed on the weights of the linear components, the weights can be marginalized out with an operation in pure linear algebra, which can (often) be made fast. We illustrate this model by demonstrating the applicability of a linear model for the non-linear systematics in K2 time-series data, where the dominant noise source for many stars is spacecraft motion and variability.

Markov Chain Monte Carlo (MCMC) methods for sampling probability density functions (combined with abundant computational resources) have transformed the sciences, especially in performing probabilistic inferences, or fitting models to data. In this primarily pedagogical contribution, we give a brief overview of the most basic MCMC method and some practical advice for the use of MCMC in real inference problems. We give advice on method choice, tuning for performance, methods for initialization, tests of convergence, troubleshooting, and use of the chain output to produce or report parameter estimates with associated uncertainties. We argue that autocorrelation time is the most important test for convergence, as it directly connects to the uncertainty on the sampling estimate of any quantity of interest. We emphasize that sampling is a method for doing integrals; this guides our thinking about how MCMC output is best used.

We describe a theoretical approach to finding spontaneously symmetry-broken electronic phases due to strong electronic interactions when using recently developed slave-particle (slave-boson) approaches based on occupation numbers. We describe why, to date, spontaneous symmetry breaking has proven difficult to achieve in such approaches. We then provide a total energy based approach for introducing auxiliary symmetry-breaking fields into the solution of the slave-particle problem that leads to lowered total energies for symmetry-broken phases. We point out that not all slave-particle approaches yield energy lowering: the slave-particle model being used must explicitly describe the degrees of freedom that break symmetry. Finally, our total energy approach permits us to greatly simplify the formalism used to achieve a self-consistent solution between spinon and slave modes while increasing the numerical stability and greatly speeding up the calculations.

Determining the three dimensional arrangement of proteins in a complex is highly beneficial for uncovering mechanistic function and interpreting genetic variation in coding genes comprising protein complexes. There are several methods for determining co-complex interactions between proteins, among them co-fractionation / mass spectrometry (CF-MS), but it remains difficult to identify directly contacting subunits within a multi-protein complex. Correlation analysis of CF-MS profiles shows promise in detecting protein complexes as a whole but is limited in its ability to infer direct physical contacts among proteins in sub-complexes. To identify direct protein-protein contacts within human protein complexes we learn a sparse conditional dependency graph from approximately 3,000 CF-MS experiments on human cell lines. We show substantial performance gains in estimating direct interactions compared to correlation analysis on a benchmark of large protein complexes with solved three-dimensional structures. We demonstrate the method's value in determining the three dimensional arrangement of proteins by making predictions for complexes without known structure (the exocyst and tRNA multi-synthetase complex) and by establishing evidence for the structural position of a recently discovered component of the core human EKC/KEOPS complex, GON7/C14ORF142, providing a more complete 3D model of the complex. Direct contact prediction provides easily calculable additional structural information for large-scale protein complex mapping studies and should be broadly applicable across organisms as more CF-MS datasets become available.

Techniques for approximately contracting tensor networks are limited in how efficiently they can make use of parallel computing resources. In this work we demonstrate and characterize a Monte Carlo approach to the tensor network renormalization group method which can be used straightforwardly on modern computing architectures. We demonstrate the efficiency of the technique and show that Monte Carlo tensor network renormalization provides an attractive path to improving the accuracy of a wide class of challenging computations while also providing useful estimates of uncertainty and a statistical guarantee of unbiased results.

Upcoming cosmic microwave background (CMB) surveys will soon make the first detection of the polarized Sunyaev-Zel'dovich effect, the linear polarization generated by the scattering of CMB photons on the free electrons present in collapsed objects. Measurement of this polarization along with knowledge of the electron density of the objects allows a determination of the quadrupolar temperature anisotropy of the CMB as viewed from the space-time location of the objects. Maps of these remote temperature quadrupoles have several cosmological applications. Here we propose a new application: reconstruction of the cosmological reionization history. We show that with quadrupole measurements out to redshift 3, constraints on the mean optical depth can be improved by an order of magnitude beyond the CMB cosmic variance limit.

Optical imaging methods using calcium indicators are critical for monitoring the activity of large neuronal populations in vivo. Imaging experiments typically generate a large amount of data that needs to be processed to extract the activity of the imaged neuronal sources. While deriving such processing algorithms is an active area of research, most existing methods require the processing of large amounts of data at a time, rendering them vulnerable to the volume of the recorded data, and preventing real-time experimental interrogation. Here we introduce OnACID, an Online framework for the Analysis of streaming Calcium Imaging Data, including i) motion artifact correction, ii) neuronal source extraction, and iii) activity denoising and deconvolution. Our approach combines and extends previous work on online dictionary learning and calcium imaging data analysis, to deliver an automated pipeline that can discover and track the activity of hundreds of cells in real time, thereby enabling new types of closed-loop experiments. We apply our algorithm on two large scale experimental datasets, benchmark its performance on manually annotated data, and show that it outperforms a popular offline approach.

We determine the physical properties of a sample of SMGs in the COSMOS field that were pre-selected at the observed wavelength of λobs=1.1 mm, and followed up at λobs=1.3 mm with ALMA. We used MAGPHYS to fit the panchromatic (ultraviolet to radio) SEDs of 124 of the target SMGs, 19.4% of which are spectroscopically confirmed. The SED analysis was complemented by estimating the gas masses of the SMGs by using the λobs=1.3 mm emission as a tracer of the molecular gas. The sample median and 16th-84th percentile ranges of the stellar masses, SFRs, dust temperatures, and dust and gas masses were derived to be log(M⋆/M⊙)=11.09+0.41−0.53, SFR=402+661−233 M⊙ yr−1, Tdust=39.7+9.7−7.4 K, log(Mdust/M⊙)=9.01+0.20−0.31, and log(Mgas/M⊙)=11.34+0.20−0.23, respectively. The median gas-to-dust ratio and gas fraction were found to be 120+73−30 and 0.62+0.27−0.23, respectively. We found that 57.3% of our SMGs populate the main sequence (MS) of star-forming galaxies, while 41.9% of the sources lie above the MS by a factor of >3 (one source lies below the MS). The largest 3 GHz radio sizes are found among the MS sources. Those SMGs that appear irregular in the rest-frame UV are predominantly starbursts, while the MS SMGs are mostly disk-like. The larger radio-emitting sizes of the MS SMGs compared to starbursts is a likely indication of their more widespread, less intense star formation. The irregular UV morphologies of the starburst SMGs are likely to echo their merger nature. Our results suggest that the transition from high-z SMGs to local ellipticals via compact, quiescent galaxies (cQGs) at z∼2 might not be universal, and the latter population might also descend from the so-called blue nuggets.

We analyze heating and cooling processes in an idealized simulation of a cool-core cluster, where momentum-driven AGN feedback balances radiative cooling in a time-averaged sense. We find that, on average, energy dissipation via shock waves is almost an order of magnitude higher than via turbulence. Most of the shock waves in the simulation are very weak shocks with Mach numbers smaller than 1.5, but the stronger shocks, although rare, dissipate energy more effectively. We find that shock dissipation is a steep function of radius, with most of the energy dissipated within 30 kpc, while radiative cooling loses area less concentrated. However, adiabatic processes and mixing (of post-shock materials and the surrounding gas) are able to redistribute the heat throughout the core. A considerable fraction of the AGN energy also escapes the core region. The cluster goes through cycles of AGN outbursts accompanied by periods of enhanced precipitation and star formation, over Gyr timescales. The cluster core is under-heated at the end of each cycle, but over-heated at the peak of the AGN outburst. During the heating-dominant phase, turbulent dissipation alone is often able to balance radiative cooling at every radius but, when this is occurs, shock waves inevitably dissipate even more energy. Our simulation explains why some clusters, such as Abell 2029, are cooling dominated, while in some other clusters, such as Perseus, various heating mechanisms including shock heating, turbulent dissipation and bubble mixing can all individually balance cooling, and together, overheat the core.