Search

We present a new, fully generative model for constructing astronomical catalogs from optical telescope image sets. Each pixel intensity is treated as a random variable with parameters that depend on the latent properties of stars and galaxies. These latent properties are themselves modeled as random. We compare two procedures for posterior inference. One procedure is based on Markov chain Monte Carlo (MCMC) while the other is based on variational inference (VI). The MCMC procedure excels at quantifying uncertainty, while the VI procedure is 1000 times faster. On a supercomputer, the VI procedure efficiently uses 665,000 CPU cores to construct an astronomical catalog from 50 terabytes of images in 14.6 minutes, demonstrating the scaling characteristics necessary to construct catalogs for upcoming astronomical surveys.

A key task in astronomy is to locate astronomical objects in images and to characterize them according to physical parameters such as brightness, color, and morphology. This task, known as cataloging, is challenging for several reasons: many astronomical objects are much dimmer than the sky background, labeled data is generally unavailable, overlapping astronomical objects must be resolved collectively, and the datasets are enormous – terabytes now, petabytes soon. In this work, we infer an astronomical catalog from 55 TB of imaging data using Celeste, a Bayesian variational inference code written entirely in the high-productivity programming language Julia. Using over 1.3 million threads on 650,000 Intel Xeon Phi cores of the Cori Phase II supercomputer, Celeste achieves a peak rate of 1.54 DP PFLOP/s. Celeste is able to jointly optimize parameters for 188 M stars and galaxies, loading and processing 178 TB across 8192 nodes in 14.6 min. To achieve this, Celeste exploits parallelism at multiple levels (cluster, node, and thread) and accelerates I/O through Cori's burst buffer. Julia's native performance enables Celeste to employ high-level constructs without resorting to hand-written or generated low-level code (C/C++/Fortran) while still achieving petascale performance.

The first supramolecular cage formed by three benzo-15-crown-5 macrocycles encapsulating a [Dy(OH2)8]3+ guest cation is reported, with the Dy(iii) centre exhibiting local pseudo square antiprismatic D4d symmetry. The anisotropy barrier extracted from ac susceptibility studies, emission spectroscopy and ab initio calculations reveals that the second excited state Kramers doublet plays a key role in the magnetization dynamics due to the Ising character and near coparallel nature of the ground and first excited Kramers doublets. [ABSTRACT FROM AUTHOR]

Abstract Background Researchers conducting molecular phylogenetic studies are frequently faced with the decision of what to do when weak branch support is obtained for key nodes of importance. As one solution, the researcher may choose to sequence additional orthologous genes of appropriate evolutionary rate for the taxa in the study. However, generating large, complete data matrices can become increasingly difficult as the number of characters increases. A few empirical studies have shown that augmenting genes even for a subset of taxa can improve branch support. However, because each study differs in the number of characters and taxa, there is still a need for additional studies that examine whether incomplete sampling designs are likely to aid at increasing deep node resolution. We target Gracillariidae, a Cretaceous-age (~100 Ma) group of leaf-mining moths to test whether the strategy of adding genes for a subset of taxa can improve branch support for deep nodes. We initially sequenced ten genes (8,418 bp) for 57 taxa that represent the major lineages of Gracillariidae plus outgroups. After finding that many deep divergences remained weakly supported, we sequenced eleven additional genes (6,375 bp) for a 27-taxon subset. We then compared results from different data sets to assess whether one sampling design can be favored over another. The concatenated data set comprising all genes and all taxa and three other data sets of different taxon and gene sub-sampling design were analyzed with maximum likelihood. Each data set was subject to five different models and partitioning schemes of non-synonymous and synonymous changes. Statistical significance of non-monophyly was examined with the Approximately Unbiased (AU) test. Results Partial augmentation of genes led to high support for deep divergences, especially when non-synonymous changes were analyzed alone. Increasing the number of taxa without an increase in number of characters led to lower bootstrap support; increasing the number of characters without increasing the number of taxa generally increased bootstrap support. More than three-quarters of nodes were supported with bootstrap values greater than 80% when all taxa and genes were combined. Gracillariidae, Lithocolletinae + Leucanthiza, and Acrocercops and Parectopa groups were strongly supported in nearly every analysis. Gracillaria group was well supported in some analyses, but less so in others. We find strong evidence for the exclusion of Douglasiidae from Gracillarioidea sensu Davis and Robinson (1998). Our results strongly support the monophyly of a G.B.R.Y. clade, a group comprised of Gracillariidae + Bucculatricidae + Roeslerstammiidae + Yponomeutidae, when analyzed with non-synonymous changes only, but this group was frequently split when synonymous and non-synonymous substitutions were analyzed together. Conclusions 1) Partially or fully augmenting a data set with more characters increased bootstrap support for particular deep nodes, and this increase was dramatic when non-synonymous changes were analyzed alone. Thus, the addition of sites that have low levels of saturation and compositional heterogeneity can greatly improve results. 2) Gracillarioidea, as defined by Davis and Robinson (1998), clearly do not include Douglasiidae, and changes to current classification will be required. 3) Gracillariidae were monophyletic in all analyses conducted, and nearly all species can be placed into one of six strongly supported clades though relationships among these remain unclear. 4) The difficulty in determining the phylogenetic placement of Bucculatricidae is probably attributable to compositional heterogeneity at the third codon position. From our tests for compositional heterogeneity and strong bootstrap values obtained when synonymous changes are excluded, we tentatively conclude that Bucculatricidae is closely related to Gracillariidae + Roeslerstammiidae + Yponomeutidae.

Abstract Background In the mega-diverse insect order Lepidoptera (butterflies and moths; 165,000 described species), deeper relationships are little understood within the clade Ditrysia, to which 98% of the species belong. To begin addressing this problem, we tested the ability of five protein-coding nuclear genes (6.7 kb total), and character subsets therein, to resolve relationships among 123 species representing 27 (of 33) superfamilies and 55 (of 100) families of Ditrysia under maximum likelihood analysis. Results Our trees show broad concordance with previous morphological hypotheses of ditrysian phylogeny, although most relationships among superfamilies are weakly supported. There are also notable surprises, such as a consistently closer relationship of Pyraloidea than of butterflies to most Macrolepidoptera. Monophyly is significantly rejected by one or more character sets for the putative clades Macrolepidoptera as currently defined (P < 0.05) and Macrolepidoptera excluding Noctuoidea and Bombycoidea sensu lato (P ≤ 0.005), and nearly so for the superfamily Drepanoidea as currently defined (P < 0.08). Superfamilies are typically recovered or nearly so, but usually without strong support. Relationships within superfamilies and families, however, are often robustly resolved. We provide some of the first strong molecular evidence on deeper splits within Pyraloidea, Tortricoidea, Geometroidea, Noctuoidea and others. Separate analyses of mostly synonymous versus non-synonymous character sets revealed notable differences (though not strong conflict), including a marked influence of compositional heterogeneity on apparent signal in the third codon position (nt3). As available model partitioning methods cannot correct for this variation, we assessed overall phylogeny resolution through separate examination of trees from each character set. Exploration of "tree space" with GARLI, using grid computing, showed that hundreds of searches are typically needed to find the best-feasible phylogeny estimate for these data. Conclusion Our results (a) corroborate the broad outlines of the current working phylogenetic hypothesis for Ditrysia, (b) demonstrate that some prominent features of that hypothesis, including the position of the butterflies, need revision, and (c) resolve the majority of family and subfamily relationships within superfamilies as thus far sampled. Much further gene and taxon sampling will be needed, however, to strongly resolve individual deeper nodes.

The reaction of exo -glycals with azidotrimethylsilane in the presence of a Brønsted acid leads to the generation of the corresponding C,N-glycosyl azides. The majority of these glycosylation reactions proceed at room temperature with short reaction times. In addition, the targeted products were obtained in high yields with exclusive diastereoselectivity to the α-anomer in pyranose-based derivatives. Carbohydrate units based on mannose, galactose, arabinose, and ribose were also shown to proceed in high yields.

Detecting differentially expressed genes is important for characterizing subpopulations of cells. In scRNA-seq data, however, nuisance variation due to technical factors like sequencing depth and RNA capture efficiency obscures the underlying biological signal. Deep generative models have been extensively applied to scRNA-seq data, with a special focus on embedding cells into a low-dimensional latent space and correcting for batch effects. However, little attention has been paid to the problem of utilizing the uncertainty from the deep generative model for differential expression (DE). Furthermore, the existing approaches do not allow for controlling for effect size or the false discovery rate (FDR). Here, we present lvm-DE, a generic Bayesian approach for performing DE predictions from a fitted deep generative model, while controlling the FDR. We apply the lvm-DE framework to scVI and scSphere, two deep generative models. The resulting approaches outperform state-of-the-art methods at estimating the log fold change in gene expression levels as well as detecting differentially expressed genes between subpopulations of cells.

A novel methodology for the arylation of exo -glycals has been developed. A range of exo -glycals underwent reactions with aryl iodides in the presence of a palladium catalyst. The transformation proceeded in a stereoselective manner to afford Z -isomers. The developed transformation demonstrated excellent functional group tolerance.

The paired measurement of RNA and surface proteins in single cells with cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) is a promising approach to connect transcriptional variation with cell phenotypes and functions. However, combining these paired views into a unified representation of cell state is made challenging by the unique technical characteristics of each measurement. Here we present Total Variational Inference (totalVI; https://scvi-tools.org ), a framework for end-to-end joint analysis of CITE-seq data that probabilistically represents the data as a composite of biological and technical factors, including protein background and batch effects. To evaluate totalVI's performance, we profiled immune cells from murine spleen and lymph nodes with CITE-seq, measuring over 100 surface proteins. We demonstrate that totalVI provides a cohesive solution for common analysis tasks such as dimensionality reduction, the integration of datasets with different measured proteins, estimation of correlations between molecules and differential expression testing.

As the number of single-cell transcriptomics datasets grows, the natural next step is to integrate the accumulating data to achieve a common ontology of cell types and states. However, it is not straightforward to compare gene expression levels across datasets and to automatically assign cell type labels in a new dataset based on existing annotations. In this manuscript, we demonstrate that our previously developed method, scVI, provides an effective and fully probabilistic approach for joint representation and analysis of scRNA-seq data, while accounting for uncertainty caused by biological and measurement noise. We also introduce single-cell ANnotation using Variational Inference (scANVI), a semi-supervised variant of scVI designed to leverage existing cell state annotations. We demonstrate that scVI and scANVI compare favorably to state-of-the-art methods for data integration and cell state annotation in terms of accuracy, scalability, and adaptability to challenging settings. In contrast to existing methods, scVI and scANVI integrate multiple datasets with a single generative model that can be directly used for downstream tasks, such as differential expression. Both methods are easily accessible through scvi-tools., (© 2021 The Authors. Published under the terms of the CC BY 4.0 license.)

A tridentate ligand LH 3 (C 11 H 13 N 3 O 4 ) comprising o-vanillin, hydrazone and oxime donor groups has been employed to prepare a new class of di- and tetranuclear Ln III complexes. The reaction of LH 3 with Ln(NO 3 ) 3 ·xH 2 O in the presence of a suitable base yields the dinuclear Dy III complex [Dy 2 (LH 2 ) 4 (CH 3 OH)][NO 3 ] 2 (1) and the tetranuclear complexes [Dy 4 (LH) 4 (LH 2 ) 2 (OH) 2 ]·2H 2 O (2) and [Gd 4 (LH) 4 (LH 2 ) 2 (OMe) 2 ]·nH 2 O, (3). In these complexes, LH 3 is either monodeprotonated (1) or a mixture of mono- and doubly-deprotonated ligands (2 and 3) binding lanthanide ions via N diazine , O hydrazone , and O vanillin donors, while the remaining vacant coordination sites are occupied by O MeOH (1), O hydroxide (2) and methoxides (3). DC magnetic susceptibility studies on the isotropic tetranuclear Gd III complex (3) reveal weak antiferromagnetic exchange interactions between the Ln III ions. AC studies reveal that the dinuclear complex (1) exhibits field-induced slow relaxation of magnetization with U eff = 43.4 K, whereas 2 is a single molecule magnet, exhibiting slow relaxation of magnetization under zero field below 18 K, which is modelled using a combination of Orbach (U eff /k B = 26.7 K) and Raman relaxation processes.

Single-cell transcriptome measurements can reveal unexplored biological diversity, but they suffer from technical noise and bias that must be modeled to account for the resulting uncertainty in downstream analyses. Here we introduce single-cell variational inference (scVI), a ready-to-use scalable framework for the probabilistic representation and analysis of gene expression in single cells ( https://github.com/YosefLab/scVI ). scVI uses stochastic optimization and deep neural networks to aggregate information across similar cells and genes and to approximate the distributions that underlie observed expression values, while accounting for batch effects and limited sensitivity. We used scVI for a range of fundamental analysis tasks including batch correction, visualization, clustering, and differential expression, and achieved high accuracy for each task.

Two complexes comprising 9-coordinate capped square antiprismatic [Ln(NO 3 ) 3 (OH 2 ) 2 (MeOH)] units [Ln(III) = Dy 6; Tb 7] are reported in which the metal complexes are hydrogen-bonded to 15C5 (15-crown-5) macrocycles to form supramolecular chains, {[Ln(NO 3 ) 3 (OH 2 ) 2 (MeOH)]·(15C5)} n . Alternating current magnetic susceptibility measurements supported by ab initio studies show field-induced SMM (single-molecule magnet) behavior for 6, but rapid relaxation of the magnetization for 7 because of the presence of dominant quantum tunneling processes as evidenced by the presence of a significant calculated tunnel splitting within the ground-state multiplet. Modeling the high-resolution emission spectra for 6 afforded energies of 37 ± 5 and 28 ± 5 cm -1 for the first-excited-state Stark sublevels of the two crystallographically independent Dy1 and Dy2 ions, in excellent agreement with the calculated values of 31 and 21 cm -1 for ΔE 1 derived from ab initio studies.

The introduction of the Schiff base ligand N-salicylidene-2-amino-5-chlorobenzoic acid (sacbH 2 ) in 4f-metal chemistry has afforded a new dinuclear complex, [Dy 2 (NO 3 ) 4 (sacbH) 2 (H 2 O) 2 (MeCN) 2 ] (1), with the metal ions adopting a rare spherical tricapped trigonal prismatic coordination geometry. The deprotonated phenoxido O atoms of the organic chelate occupy the axial triangular faces of the prism and were found to be very close to the main anisotropy axes of the two Dy III ions. As a result, the {Dy III 2 } compound exhibits frequency- and temperature-dependent out-of-phase ac signals below ∼25 K in the absence of a static dc field, yielding an energy barrier of 109.3(1) K for the reversal of magnetization. Fast and efficient quantum tunneling of magnetization, attributed to the strong tails of signals below ∼15 K, was suppressed through the application of a small dc field, yielding entirely visible χ M ″ signals below 27 K. Single-crystal magnetic hysteresis studies confirmed the single-molecule magnet (SMM) behavior of 1; the hysteresis loops appear at temperatures below ∼5 K, which is one of the highest blocking temperatures in the field of 4f-SMMs to date. This joint magneto-structural and ab initio study demonstrates the ability of more common coordination numbers (i.e., 9), but with rare coordination geometries (i.e., spherical tricapped trigonal prismatic), to promote axiality that enhances the molecular anisotropy and subsequently the magnetization dynamics of the system.

Johannes Kepler (1571-1630) makes extensive use of souls and spiritus in his natural philosophy. Recent studies have highlighted their importance in his accounts of celestial generation and astrology. In this study, I would like to address two pressing issues. The first is Kepler's context. The biological side of his natural philosophy is not naively Aristotelian. Instead, he is up to date with contemporary discussions in medically flavored natural philosophy. I will examine his relationship to Melanchthon's anatomical-theological Liber de anima (1552) and to Jean Femel's very popular Physiologia (1567), two Galenic sources with a noticeable impact on how he understands the functions of life. The other issue that will direct my article is force at a distance. Medical ideas deeply inform Kepler's theories of light and solar force (virtus motrix). It will become clear that they are not a hindrance even to the hardcore of his celestial physics. Instead, he makes use of soul and spiritus in order to develop a fully mathematized dynamics.

Robust resolution of controversial higher-level groupings within Arthropoda requires additional sources of characters. Toward this end, elongation factor-2 sequences (1899 nucleotides) were generated from 17 arthropod taxa (5 chelicerates, 6 crustaceans, 3 hexapods, 3 myriapods) plus an onychophoran and a tardigrade as outgroups. Likelihood and parsimony analyses of nucleotide and amino acid data sets consistently recovered Myriapoda and major chelicerate groups with high bootstrap support. Crustacea + Hexapoda (= Pancrustacea) was recovered with moderate support, whereas the conflicting group Myriapoda + Hexapoda (= Atelocerata) was never recovered and bootstrap values were always <5%. With additional nonarthropod sequences included, one indel supports monophyly of Tardigrada, Onychophora, and Arthropoda relative to molluscan, annelidan, and mammalian outgroups. New and previously published sequences from RNA polymerase II (1038 nucleotides) and elongation factor-1alpha (1092 nucleotides) were analyzed for the same taxa. A comparison of bootstrap values from the three genes analyzed separately revealed widely varying values for some clades, although there was never strong support for conflicting groups. In combined analyses, there was strong bootstrap support for the generally accepted clades Arachnida, Arthropoda, Euchelicerata, Hexapoda, and Pycnogonida, and for Chelicerata, Myriapoda, and Pancrustacea, whose monophyly is more controversial. Recovery of some additional groups was fairly robust to method of analysis but bootstrap values were not high; these included Pancrustacea + Chelicerata, Hexapoda + Cephalocarida + Remipedia, Cephalocarida + Remipedia, and Malaocostraca + Cirripedia. Atelocerata (= Myriapoda + Hexapoda) was never recovered. Elongation factor-2 is now the second protein-encoding, nuclear gene (in addition to RNA polymerase II) to support Pancrustacea over Atelocerata. Atelocerata is widely cited in morphology-based analyses, and the discrepancy between results derived from molecular and morphological data deserves greater attention., (Copyright 2001 Academic Press.)

Phylogenetic utility for the nuclear gene encoding dopa decarboxylase (DDC), little used in systematics, was recently demonstrated within the noctuid moth subfamily Heliothinae. Here we extend the test of the utility of a 709-bp DDC fragment to deeper levels, analyzing 49 species representing major groups across the superfamily Noctuoidea. Parsimony, distance, and maximum-likelihood analyses recover all or nearly all of a set of "test clades" supported by clear morphological synapomorphies, spanning a wide range of taxonomic levels. DDC also upholds a recent proposal that the Noctuidae are paraphyletic. Nt3 contributes a majority of the signal and recovers the basal split between Notodontidae and all other noctuoids, despite a plateau of nt3 divergence at this level. However, nonsynonymous changes also support groups at all levels, and in contrast to nt3, amino acid divergence shows no plateau. The utility of DDC promises to extend back to the early Tertiary and Cretaceous, a time span for which few suitable genes have been identified., (Copyright 2000 Academic Press.)

A central question concerning data collection strategy for molecular phylogenies has been, is it better to increase the number of characters or the number of taxa sampled to improve the robustness of a phylogeny estimate? A recent simulation study concluded that increasing the number of taxa sampled is preferable to increasing the number of nucleotide characters, if taxa are chosen specifically to break up long branches. We explore this hypothesis by using empirical data from noctuoid moths, one of the largest superfamilies of insects. Separate studies of two nuclear genes, elongation factor-1 alpha (EF-1 alpha) and dopa decarboxylase (DDC), have yielded similar gene trees and high concordance with morphological groupings for 49 exemplar species. However, support levels were quite low for nodes deeper than the subfamily level. We tested the effects on phylogenetic signal of (1) increasing the taxon sampling by nearly 60%, to 77 species, and (2) combining data from the two genes in a single analysis. Surprisingly, the increased taxon sampling, although designed to break up long branches, generated greater disagreement between the two gene data sets and decreased support levels for deeper nodes. We appear to have inadvertently introduced new long branches, and breaking these up may require a yet larger taxon sample. Sampling additional characters (combining data) greatly increased the phylogenetic signal. To contrast the potential effect of combining data from independent genes with collection of the same total number of characters from a single gene, we simulated the latter by bootstrap augmentation of the single-gene data sets. Support levels for combined data were at least as high as those for the bootstrap-augmented data set for DDC and were much higher than those for the augmented EF-1 alpha data set. This supports the view that in obtaining additional sequence data to solve a refractory systematic problem, it is prudent to take them from an independent gene.

Recent phylogenetic analyses using molecular data suggest that hexapods are more closely related to crustaceans than to myriapods, a result that conflicts with long-held morphology-based hypotheses. Here we contribute additional information to this debate by conducting phylogenetic analyses on two nuclear protein-encoding genes, elongation factor-1 alpha (EF-1 alpha) and the largest subunit of RNA polymerase II (Pol II), from an extensive sample of arthropod taxa. Results were obtained from two data sets. One data set comprised 1092 nucleotides (364 amino acids) of EF-1 alpha and 372 nucleotides (124 amino acids) of Pol II from 30 arthropods and three lobopods. The other data set contained the same EF-1 alpha fragment and an expanded 1038-nucleotide (346-amino-acid) sample of Pol II from 17 arthropod taxa. Results from maximum-parsimony and maximum-likelihood analyses strongly supported the existence of a Crustacea + Hexapoda clade (Pancrustacea) over a Myriapoda + Hexapoda clade (Atelocerata). The apparent incompatibility between the molecule-based Pancrustacea hypothesis and morphology-based Atelocerata hypothesis is discussed.

Compared to the number of genes available for study of both younger and older divergences, few genes have yet been identified that can strongly resolve phylogenetic splits of Mesozoic age ( approximately 65-250 mya). Thus, reconstruction of Mesozoic-age phylogenies, exemplified by basal divergences within the major orders of holometabolous insects, is likely to be especially dependent on combining multiple lines of evidence. This study tests the potential of the 18S ribosomal RNA gene for reconstructing Mesozoic-aged divergences within the insect order Lepidoptera and its ability when combined with a second, previously analyzed nuclear gene (phosphoenolpyruvate carboxykinase, PEPCK) to strongly resolve these relationships. 18S sequences were obtained for 21 taxa, representing major clades of Lepidoptera plus outgroups from the other "panorpoid orders. A well-corroborated morphology-based "test phylogeny was used to evaluate the effects of partitioning the 18S gene according to variable versus conserved domains, paired versus unpaired sites in the secondary structure, and transition versus transversion substitutions. Likelihood and unweighted parsimony analyses of the 18S data recover the "test phylogeny" almost completely, with no improvement of agreement or support provided by any form of weighting or partitioning. No conflict in signal between 18S and PEPCK was detected by the partition homogeneity test. Combined parsimony analysis yielded strong bootstrap support for nearly all relationships, much higher than for either gene alone, thereby also providing strong evidence on several hypotheses about the early evolution of lepidopteran-plant interactions. These genes in combination may be widely useful for resolving insect divergences of comparable age., (Copyright 2000 Academic Press.)

To extend initial characterizations of their phylogenetic utility, sequences from the nuclear genes for elongation factor-1 alpha (EF-1 alpha) and dopa decarboxylase (DDC) are tested for phylogenetic concordance with each other and with previous morphological evidence within the giant silk moth tribe Attacini (Lepidoptera: Saturniidae). The sampling of DDC is expanded from the 690 basepairs of previous studies to 1051 basepairs in the current study. All nine attacine genera are sampled. EF-1 alpha and DDC agree in the placement of seven of nine genera, with placement of the other two not in strong conflict. Combination of the gene sequences results in a nearly fully resolved tree that is consistent with EF-1 alpha alone and agrees with morphology in five of eight groups. Conflict between molecules and morphology is confined to deeper-level relationships within Attacini, where node support for the molecular hypotheses, but not the morphological hypotheses, is generally very strong. A strong signal is contributed by synonymous substitutions in both genes, and by nonsynonymous change particularly in DDC. The molecular phylogeny supports a revision of attacine biogeography in that neither East Asian nor New World genera form monophyletic groups.