Payel Das, Kahini Wadhawan, Oscar Chang, Tom Sercu, Cicero Dos Santos, Matthew Riemer, Vijil Chenthamarakshan, Inkit Padhi, Aleksandra Mojsilovic
Oct 17, 2018·q-bio.QM·PDF Given the emerging global threat of antimicrobial resistance, new methods for next-generation antimicrobial design are urgently needed. We report a peptide generation framework PepCVAE, based on a semi-supervised variational autoencoder (VAE) model, for designing novel antimicrobial peptide (AMP) sequences. Our model learns a rich latent space of the biological peptide context by taking advantage of abundant, unlabeled peptide sequences. The model further learns a disentangled antimicrobial attribute space by using the feedback from a jointly trained AMP classifier that uses limited labeled instances. The disentangled representation allows for controllable generation of AMPs. Extensive analysis of the PepCVAE-generated sequences reveals superior performance of our model in comparison to a plain VAE, as PepCVAE generates novel AMP sequences with higher long-range diversity, while being closer to the training distribution of biological peptides. These features are highly desired in next-generation antimicrobial design.
Payel Das, Keith Hawkins, Paula Jofre
Mar 22, 2019·astro-ph.GA·PDF We exploit the [Mg/Mn]-[Al/Fe] chemical abundance plane to help identify nearby halo stars in the 14th data release from the APOGEE survey that have been accreted on to the Milky Way. Applying a Gaussian Mixture Model, we find a `blob' of 856 likely accreted stars, with a low disc contamination rate of ~7%. Cross-matching the sample with the second data release from Gaia gives us access to parallaxes and apparent magnitudes, which place constraints on distances and intrinsic luminosities. Using a Bayesian isochrone pipeline, this enables us to estimate new ages for the accreted stars, with typical uncertainties of ~20%. Our new catalogue is further supplemented with estimates of orbital parameters. The blob stars span a metallicities between -0.5 to -2.5, and [Mg/Fe] between -0.1 to 0.5. They constitute ~30% of the metal-poor ([Fe/H] < -0.8) halo at metallicities of ~-1.4. Our new ages are mainly range between 8 to 13 Gyr, with the oldest stars the metal-poorest, and with the highest [Mg/Fe] abundance. If the blob stars are assumed to belong to a single progenitor, the ages imply that the system merged with our Milky Way around 8 Gyr ago and that star formation proceeded for ~5 Gyr. Dynamical arguments suggest that such a single progenitor would have a total mass of ~1011Msun, similar to that found by other authors using chemical evolution models and simulations. Comparing the scatter in the [Mg/Fe]-[Fe/H] plane of the blob stars to that measured for stars belonging to the Large Magellanic Cloud suggests that the blob does indeed contain stars from only one progenitor.
Yair Schiff, Brian Quanz, Payel Das, Pin-Yu Chen
The field of Deep Learning is rich with empirical evidence of human-like performance on a variety of prediction tasks. However, despite these successes, the recent Predicting Generalization in Deep Learning (PGDL) NeurIPS 2020 competition suggests that there is a need for more robust and efficient measures of network generalization. In this work, we propose a new framework for evaluating the generalization capabilities of trained networks. We use perturbation response (PR) curves that capture the accuracy change of a given network as a function of varying levels of training sample perturbation. From these PR curves, we derive novel statistics that capture generalization capability. Specifically, we introduce two new measures for accurately predicting generalization gaps: the Gi-score and Pal-score, which are inspired by the Gini coefficient and Palma ratio (measures of income inequality), that accurately predict generalization gaps. Using our framework applied to intra and inter-class sample mixup, we attain better predictive scores than the current state-of-the-art measures on a majority of tasks in the PGDL competition. In addition, we show that our framework and the proposed statistics can be used to capture to what extent a trained network is invariant to a given parametric input transformation, such as rotation or translation. Therefore, these generalization gap prediction statistics also provide a useful means for selecting optimal network architectures and hyperparameters that are invariant to a certain perturbation.
Tejas Dharamsi, Payel Das, Tejaswini Pedapati, Gregory Bramble, Vinod Muthusamy, Horst Samulowitz, Kush R. Varshney, Yuvaraj Rajamanickam, John Thomas, Justin Dauwels
Electroencephalography (EEG) is an extensively-used and well-studied technique in the field of medical diagnostics and treatment for brain disorders, including epilepsy, migraines, and tumors. The analysis and interpretation of EEGs require physicians to have specialized training, which is not common even among most doctors in the developed world, let alone the developing world where physician shortages plague society. This problem can be addressed by teleEEG that uses remote EEG analysis by experts or by local computer processing of EEGs. However, both of these options are prohibitively expensive and the second option requires abundant computing resources and infrastructure, which is another concern in developing countries where there are resource constraints on capital and computing infrastructure. In this work, we present a cloud-based deep neural network approach to provide decision support for non-specialist physicians in EEG analysis and interpretation. Named `neurology-as-a-service,' the approach requires almost no manual intervention in feature engineering and in the selection of an optimal architecture and hyperparameters of the neural network. In this study, we deploy a pipeline that includes moving EEG data to the cloud and getting optimal models for various classification tasks. Our initial prototype has been tested only in developed world environments to-date, but our intention is to test it in developing world environments in future work. We demonstrate the performance of our proposed approach using the BCI2000 EEG MMI dataset, on which our service attains 63.4% accuracy for the task of classifying real vs. imaginary activity performed by the subject, which is significantly higher than what is obtained with a shallow approach such as support vector machines.
Ravi Tejwani, Adam Liska, Hongyuan You, Jenna Reinen, Payel Das
Dec 21, 2017·q-bio.NC·PDF The goal of the present study is to identify autism using machine learning techniques and resting-state brain imaging data, leveraging the temporal variability of the functional connections (FC) as the only information. We estimated and compared the FC variability across brain regions between typical, healthy subjects and autistic population by analyzing brain imaging data from a world-wide multi-site database known as ABIDE (Autism Brain Imaging Data Exchange). Our analysis revealed that patients diagnosed with autism spectrum disorder (ASD) show increased FC variability in several brain regions that are associated with low FC variability in the typical brain. We then used the enhanced FC variability of brain regions as features for training machine learning models for ASD classification and achieved 65% accuracy in identification of ASD versus control subjects within the dataset. We also used node strength estimated from number of functional connections per node averaged over the whole scan as features for ASD classification.The results reveal that the dynamic FC measures outperform or are comparable with the static FC measures in predicting ASD.
Payel Das, Yang Huang, Ioana Ciuca, Francesca Fragkoudi
May 12, 2023·astro-ph.GA·PDF Phase-space data, chemistry, and ages together reveal a complex structure in the outer low-$α$ disc of the Milky Way. The age-vertical velocity dispersion profiles beyond the Solar Neighbourhood show a significant jump at 6 Gyr for stars beyond the Galactic plane. Stars older than 6 Gyr are significantly hotter than younger stars. The chemistry and age histograms reveal a bump at [Fe/H] = -0.5, [$α$/Fe] = 0.1, and an age of 7.2 Gyr in the outer disc. Finally, viewing the stars beyond 13.5 kpc in the age-metallicity plane reveals a faint streak just below this bump, towards lower metallicities at the same age. Given the uncertainty in age, we believe these features are linked and suggest a pericentric passage of a massive satellite 6 Gyr ago that heated pre-existing stars, led to a starburst in existing gas. New stars also formed from the metal-poorer infalling gas. The impulse approximation was used to characterise the interaction with a satellite, finding a mass of ~1e11 M$_{\odot}$, and a pericentric position between 12 and 16 kpc. The evidence points to an interaction with the Sagittarius dwarf galaxy, likely its first pericentric passage.
Michelle Doherty, Magda Arnaboldi, Payel Das, Ortwin Gerhard, J. Alfonso L. Aguerri, Robin Ciardullo, John J. Feldmeier, Kenneth C. Freeman, George H. Jacoby, Giuseppe Murante
May 12, 2009·astro-ph.CO·PDF We present high resolution FLAMES/VLT spectroscopy of intracluster planetary nebula (ICPN) candidates, targeting three new fields in the Virgo cluster core with surface brightness down to mu_B = 28.5. Based on the projected phase space information we separate the old and 12 newly-confirmed PNs into galaxy and intracluster components. The M87 PNs are confined to the extended stellar envelope of M87, within a projected radius of ~ 160 kpc, while the ICPNs are scattered across the whole surveyed region between M87 and M86. The velocity dispersions determined from the M87 PNs at projected radii of 60 kpc and 144 kpc show that the galaxy's velocity dispersion profile decreases in the outer halo, down to 78 +/- 25 km/s. A Jeans model for the M87 halo stars in the gravitational potential traced by the X-ray emission fits the observed velocity dispersion profile only if the stellar orbits are strongly radially anisotropic (beta ~= 0.4 at r ~= 10 kpc increasing to 0.8 at the outer edge), and if additionally the stellar halo is truncated at ~= 150 kpc average elliptical radius. From the spatial and velocity distribution of the ICPNs we infer that M87 and M86 are falling towards each other and that we may be observing them just before the first close pass. The inferred luminosity-specific PN numbers for the M87 halo and the ICL are in the range of values observed for old (> 10 Gyr) stellar populations (abridged).
Payel Das, James Binney
Mar 30, 2016·astro-ph.GA·PDF We fit an Extended Distribution Function (EDF) to K giants in the Sloan Extension for Galactic Understanding and Exploration (SEGUE) survey. These stars are detected to radii ~80 kpc and span a wide range in [Fe/H]. Our EDF, which depends on [Fe/H] in addition to actions, encodes the entanglement of metallicity with dynamics within the Galaxy's stellar halo. Our maximum-likelihood fit of the EDF to the data allows us to model the survey's selection function. The density profile of the K giants steepens with radius from a slope ~-2 to ~-4 at large radii. The halo's axis ratio increases with radius from 0.7 to almost unity. The metal-rich stars are more tightly confined in action space than the metal-poor stars and form a more flattened structure. A weak metallicity gradient ~-0.001 dex/kpc, a small gradient in the dispersion in [Fe/H] of ~0.001 dex/kpc, and a higher degree of radial anistropy in metal-richer stars result. Lognormal components with peaks at ~-1.5 and ~-2.3 are required to capture the overall metallicity distribution, suggestive of the existence of two populations of K giants. The spherical anisotropy parameter varies between 0.3 in the inner halo to isotropic in the outer halo. If the Sagittarius stream is included, a very similar model is found but with a stronger degree of radial anisotropy throughout.
Raphaël Pestourie, Youssef Mroueh, Thanh V. Nguyen, Payel Das, Steven G. Johnson
Surrogate models for partial-differential equations are widely used in the design of meta-materials to rapidly evaluate the behavior of composable components. However, the training cost of accurate surrogates by machine learning can rapidly increase with the number of variables. For photonic-device models, we find that this training becomes especially challenging as design regions grow larger than the optical wavelength. We present an active learning algorithm that reduces the number of training points by more than an order of magnitude for a neural-network surrogate model of optical-surface components compared to random samples. Results show that the surrogate evaluation is over two orders of magnitude faster than a direct solve, and we demonstrate how this can be exploited to accelerate large-scale engineering optimization.
Vijil Chenthamarakshan, Payel Das, Samuel C. Hoffman, Hendrik Strobelt, Inkit Padhi, Kar Wai Lim, Benjamin Hoover, Matteo Manica, Jannis Born, Teodoro Laino, Aleksandra Mojsilovic
The novel nature of SARS-CoV-2 calls for the development of efficient de novo drug design approaches. In this study, we propose an end-to-end framework, named CogMol (Controlled Generation of Molecules), for designing new drug-like small molecules targeting novel viral proteins with high affinity and off-target selectivity. CogMol combines adaptive pre-training of a molecular SMILES Variational Autoencoder (VAE) and an efficient multi-attribute controlled sampling scheme that uses guidance from attribute predictors trained on latent features. To generate novel and optimal drug-like molecules for unseen viral targets, CogMol leverages a protein-molecule binding affinity predictor that is trained using SMILES VAE embeddings and protein sequence embeddings learned unsupervised from a large corpus. CogMol framework is applied to three SARS-CoV-2 target proteins: main protease, receptor-binding domain of the spike protein, and non-structural protein 9 replicase. The generated candidates are novel at both molecular and chemical scaffold levels when compared to the training data. CogMol also includes insilico screening for assessing toxicity of parent molecules and their metabolites with a multi-task toxicity classifier, synthetic feasibility with a chemical retrosynthesis predictor, and target structure binding with docking simulations. Docking reveals favorable binding of generated molecules to the target protein structure, where 87-95 % of high affinity molecules showed docking free energy < -6 kcal/mol. When compared to approved drugs, the majority of designed compounds show low parent molecule and metabolite toxicity and high synthetic feasibility. In summary, CogMol handles multi-constraint design of synthesizable, low-toxic, drug-like molecules with high target specificity and selectivity, and does not need target-dependent fine-tuning of the framework or target structure information.
Payel Das, Brian Quanz, Pin-Yu Chen, Jae-wook Ahn, Dhruv Shah
Creativity, a process that generates novel and meaningful ideas, involves increased association between task-positive (control) and task-negative (default) networks in the human brain. Inspired by this seminal finding, in this study we propose a creative decoder within a deep generative framework, which involves direct modulation of the neuronal activation pattern after sampling from the learned latent space. The proposed approach is fully unsupervised and can be used off-the-shelf. Several novelty metrics and human evaluation were used to evaluate the creative capacity of the deep decoder. Our experiments on different image datasets (MNIST, FMNIST, MNIST+FMNIST, WikiArt and CelebA) reveal that atypical co-activation of highly activated and weakly activated neurons in a deep decoder promotes generation of novel and meaningful artifacts.
Payel Das, Angus Williams, James Binney
Aug 25, 2016·astro-ph.GA·PDF The distribution of Milky Way halo blue horizontal-branch (BHB) stars is examined using action-based extended distribution functions (EDFs) that describe the locations of stars in phase space, metallicity, and age. The parameters of the EDFs are fitted using stars observed in the Sloan Extension for Galactic Understanding and Exploration-II (SEGUE-II) survey that trace the phase-space kinematics and chemistry out to ~70 kpc. A maximum a posteriori probability (MAP) estimate method and a Markov Chain Monte Carlo method are applied, taking into account the selection function in positions, distance, and metallicity for the survey. The best-fit EDF declines with actions less steeply at actions characteristic of the inner halo than at the larger actions characteristic of the outer halo, and older ages are found at smaller actions than at larger actions. In real space, the radial density profile steepens smoothly from -2 at ~2 kpc to -4 in the outer halo, with an axis ratio ~0.7 throughout. There is no indication for rotation in the BHBs, although this is highly uncertain. A moderate level of radial anisotropy is detected, with $β_s$ varying from isotropic to between ~0.1 and ~0.3 in the outer halo depending on latitude. The BHB data are consistent with an age gradient of -0.03 Gyr kpc$^{-1}$, with some uncertainty in the distribution of the larger ages. These results are consistent with a scenario in which older, larger systems contribute to the inner halo, whilst the outer halo is primarily comprised of younger, smaller systems.
Payel Das, Ortwin Gerhard, Eugene Churazov, Irina Zhuravleva
Jul 29, 2010·astro-ph.CO·PDF We use a new non-parametric Bayesian approach to obtain the most probable mass distributions and circular velocity curves along with their confidence ranges, given deprojected density and temperature profiles of the hot gas surrounding X-ray bright elliptical galaxies. For a sample of six X-ray bright ellipticals, we find that all circular velocity curves are rising in the outer parts due to a combination of a rising temperature profile and a logarithmic pressure gradient that increases in magnitude. Comparing the circular velocity curves we obtain from X-rays to those obtained from dynamical models, we find that the former are often lower in the central ~10 kpc. This is probably due to a combination of: i) Non-thermal contributions of up to ~35% in the pressure (with stronger effects in NGC 4486), ii) multiple-temperature components in the hot gas, iii) incomplete kinematic spatial coverage in the dynamical models, and iv) mass profiles that are insufficiently general in the dynamical modelling. Complementing the total mass information from the X-rays with photometry and stellar population models to infer the dark matter content, we find evidence for massive dark matter haloes with dark matter mass fractions of ~35-80% at 2Re, rising to a maximum of 80-90% at the outermost radii. We also find that the six galaxies follow a Tully-Fisher relation with slope ~4 and that their circular velocities at 1Re correlate strongly with the velocity dispersion of the local environment. As a result, the galaxy luminosity at 1Re also correlates with the velocity dispersion of the environment. These relations suggest a close link between the properties of central X-ray bright elliptical galaxies and their environments (abridged).
Payel Das, Ortwin Gerhard, Roberto H. Mendez, Ana M. Teodorescu, Flavio de Lorenzi
May 17, 2011·astro-ph.CO·PDF We create dynamical models of the massive elliptical galaxy, NGC 4649, using the N-body made-to-measure code, NMAGIC, and kinematic constraints from long-slit and planetary nebula (PN) data. We explore a range of potentials based on previous determinations from X-ray observations and a dynamical model fitting globular cluster (GC) velocities and a stellar density profile. The X-ray mass distributions are similar in the central region but have varying outer slopes, while the GC mass profile is higher in the central region and on the upper end of the range further out. Our models cannot differentiate between the potentials in the central region, and therefore if non-thermal pressures or multi-phase components are present in the hot gas, they must be smaller than previously inferred. In the halo, we find that the PN velocities are sensitive tracers of the mass, preferring a less massive halo than that derived from the GC mass profile, but similar to one of the mass distributions derived from X-rays. Our results show that the GCs may form a dynamically distinct system, and that the properties of the hot gas derived from X-rays in the outer halo have considerable uncertainties that need to be better understood. Estimating the mass in stars using photometric information and a stellar population mass-to-light ratio, we infer a dark matter mass fraction in NGC 4649 of ~0.39 at 1Re (10.5 kpc) and ~0.78 at 4Re. We find that the stellar orbits are isotropic to mildly radial in the central ~6 kpc depending on the potential assumed. Further out, the orbital structure becomes slightly more radial along R and more isotropic along z, regardless of the potential assumed. In the equatorial plane, azimuthal velocity dispersions dominate over meridional velocity dispersions, implying that meridional velocity anisotropy is the mechanism for flattening the stellar system.
Yair Schiff, Vijil Chenthamarakshan, Samuel Hoffman, Karthikeyan Natesan Ramamurthy, Payel Das
Deep generative models have emerged as a powerful tool for learning useful molecular representations and designing novel molecules with desired properties, with applications in drug discovery and material design. However, most existing deep generative models are restricted due to lack of spatial information. Here we propose augmentation of deep generative models with topological data analysis (TDA) representations, known as persistence images, for robust encoding of 3D molecular geometry. We show that the TDA augmentation of a character-based Variational Auto-Encoder (VAE) outperforms state-of-the-art generative neural nets in accurately modeling the structural composition of the QM9 benchmark. Generated molecules are valid, novel, and diverse, while exhibiting distinct electronic property distribution, namely higher sample population with small HOMO-LUMO gap. These results demonstrate that TDA features indeed provide crucial geometric signal for learning abstract structures, which is non-trivial for existing generative models operating on string, graph, or 3D point sets to capture.
Payel Das, Jason Sanders
Apr 25, 2018·astro-ph.GA·PDF We present a new approach (MADE) that generates mass, age, and distance estimates of red giant stars from a combination of astrometric, photometric, and spectroscopic data. The core of the approach is a Bayesian artificial neural network (ANN) that learns from and completely replaces stellar isochrones. The ANN is trained using a sample of red giant stars with mass estimates from asteroseismology. A Bayesian isochrone pipeline uses the astrometric, photometric, spectroscopic, and asteroseismology data to determine posterior distributions for the training outputs: mass, age, and distance. Given new inputs, posterior predictive distributions for the outputs are computed, taking into account both input uncertainties, and uncertainties in the ANN parameters. We apply MADE to $\sim10\,000$ red giants in the overlap between the 14$^{\mathrm{th}}$ data release from the APO Galactic Evolution Experiment (APOGEE, Abolfathi et al. 2018) and the Tycho-Gaia astrometric solution (TGAS, Michalik et al. 2015). The ANN is able to reduce the uncertainty on mass, age, and distance estimates for training-set stars with high output uncertainties allocated through the Bayesian isochrone pipeline. The fractional uncertainties on mass are $<10\%$ and on age are between $10$ to $25\%$. Moreover, the time taken for our ANN to predict masses, ages, and distances for the entire catalogue of APOGEE-TGAS stars is of a similar order of the time taken by the Bayesian isochrone pipeline to run on a handful of stars. Our resulting catalogue clearly demonstrates the expected thick and thin disc components in the [M/H]-[$α$/M] plane, when examined by age.
N. Joseph Tatro, Pin-Yu Chen, Payel Das, Igor Melnyk, Prasanna Sattigeri, Rongjie Lai
The loss landscapes of deep neural networks are not well understood due to their high nonconvexity. Empirically, the local minima of these loss functions can be connected by a learned curve in model space, along which the loss remains nearly constant; a feature known as mode connectivity. Yet, current curve finding algorithms do not consider the influence of symmetry in the loss surface created by model weight permutations. We propose a more general framework to investigate the effect of symmetry on landscape connectivity by accounting for the weight permutations of the networks being connected. To approximate the optimal permutation, we introduce an inexpensive heuristic referred to as neuron alignment. Neuron alignment promotes similarity between the distribution of intermediate activations of models along the curve. We provide theoretical analysis establishing the benefit of alignment to mode connectivity based on this simple heuristic. We empirically verify that the permutation given by alignment is locally optimal via a proximal alternating minimization scheme. Empirically, optimizing the weight permutation is critical for efficiently learning a simple, planar, low-loss curve between networks that successfully generalizes. Our alignment method can significantly alleviate the recently identified robust loss barrier on the path connecting two adversarial robust models and find more robust and accurate models on the path.
Kar Wai Lim, Bhanushee Sharma, Payel Das, Vijil Chenthamarakshan, Jonathan S. Dordick
Sep 23, 2020·q-bio.QM·PDF Chemical toxicity prediction using machine learning is important in drug development to reduce repeated animal and human testing, thus saving cost and time. It is highly recommended that the predictions of computational toxicology models are mechanistically explainable. Current state of the art machine learning classifiers are based on deep neural networks, which tend to be complex and harder to interpret. In this paper, we apply a recently developed method named contrastive explanations method (CEM) to explain why a chemical or molecule is predicted to be toxic or not. In contrast to popular methods that provide explanations based on what features are present in the molecule, the CEM provides additional explanation on what features are missing from the molecule that is crucial for the prediction, known as the pertinent negative. The CEM does this by optimizing for the minimum perturbation to the model using a projected fast iterative shrinkage-thresholding algorithm (FISTA). We verified that the explanation from CEM matches known toxicophores and findings from other work.
Andrew Everall, Payel Das
Feb 27, 2019·astro-ph.GA·PDF Selection functions are vital for understanding the observational biases of spectroscopic surveys. With the wide variety of multi-object spectrographs currently in operation and becoming available soon, we require easily generalisable methods for determining the selection functions of these surveys. Previous work, however, has largely been focused on generating individual, tailored selection functions for every data release of each survey. Moreover, no methods for combining these selection functions to be used for joint catalogues have been developed. We have developed a Poisson likelihood estimation method for calculating selection functions in a Bayesian framework, which can be generalised to any multi-object spectrograph. We include a robust treatment of overlapping fields within a survey as well as selection functions for combined samples with overlapping footprints. We also provide a method for transforming the selection function that depends on the sky positions, colour, and apparent magnitude of a star to one that depends on the galactic location, metallicity, mass, and age of a star. This `intrinsic' selection function is invaluable for chemodynamical models of the Milky Way. We demonstrate that our method is successful at recreating synthetic spectroscopic samples selected from a mock galaxy catalogue.
Hamid Dadkhahi, Karthikeyan Shanmugam, Jesus Rios, Payel Das, Samuel Hoffman, Troy David Loeffler, Subramanian Sankaranarayanan
We consider the problem of black-box function optimization over the boolean hypercube. Despite the vast literature on black-box function optimization over continuous domains, not much attention has been paid to learning models for optimization over combinatorial domains until recently. However, the computational complexity of the recently devised algorithms are prohibitive even for moderate numbers of variables; drawing one sample using the existing algorithms is more expensive than a function evaluation for many black-box functions of interest. To address this problem, we propose a computationally efficient model learning algorithm based on multilinear polynomials and exponential weight updates. In the proposed algorithm, we alternate between simulated annealing with respect to the current polynomial representation and updating the weights using monomial experts' advice. Numerical experiments on various datasets in both unconstrained and sum-constrained boolean optimization indicate the competitive performance of the proposed algorithm, while improving the computational time up to several orders of magnitude compared to state-of-the-art algorithms in the literature.