Large subglacial lakes in East Antarctica at the onset of fast-flowing ice streams

Water plays a crucial role in ice-sheet stability and the onset of ice streams. Subglacial lake water moves between lakes and rapidly drains, causing catastrophic floods. The exact mechanisms by which subglacial lakes influence ice-sheet dynamics are unknown, however, and large subglacial lakes have not been closely associated with rapidly flowing ice streams. Here we use satellite imagery and ice-surface elevations to identify a region of subglacial lakes, similar in total area to Lake Vostok, at the onset region of the Recovery Glacier ice stream in East Antarctica and predicted by ice-sheet models. We define four lakes through extensive, flat, featureless regions of ice surface bounded by upstream troughs and downstream ridges. Using ice velocities determined using interferometric synthetic aperture radar (InSAR), we find the onset of rapid flow (moving at 20 to 30 m yr(-1)) of the tributaries to the Recovery Glacier ice stream in a 280-km-wide segment at the downslope margins of these four subglacial lakes. We conclude that the subglacial lakes initiate and maintain rapid ice flow through either active modification of the basal thermal regime of the ice sheet by lake accretion or through scouring bedrock channels in periodic drainage events. We suggest that the role of subglacial lakes needs to be considered in ice-sheet mass balance assessments.     

Epithelial-cell-intrinsic IKK-beta expression regulates intestinal immune homeostasis

Intestinal epithelial cells (IECs) provide a primary physical barrier against commensal and pathogenic microorganisms in the gastrointestinal (GI) tract, but the influence of IECs on the development and regulation of immunity to infection is unknown. Here we show that IEC-intrinsic IkappaB kinase (IKK)-beta-dependent gene expression is a critical regulator of responses of dendritic cells and CD4+ T cells in the GI tract. Mice with an IEC-specific deletion of IKK-beta show a reduced expression of the epithelial-cell-restricted cytokine thymic stromal lymphopoietin in the intestine and, after infection with the gut-dwelling parasite Trichuris, fail to develop a pathogen-specific CD4+ T helper type 2 (T(H)2) response and are unable to eradicate infection. Further, these animals show exacerbated production of dendritic-cell-derived interleukin-12/23p40 and tumour necrosis factor-alpha, increased levels of CD4+ T-cell-derived interferon-gamma and interleukin-17, and develop severe intestinal inflammation. Blockade of proinflammatory cytokines during Trichuris infection ablates the requirement for IKK-beta in IECs to promote CD4+ T(H)2 cell-dependent immunity, identifying an essential function for IECs in tissue-specific conditioning of dendritic cells and limiting type 1 cytokine production in the GI tract. These results indicate that the balance of IKK-beta-dependent gene expression in the intestinal epithelium is crucial in intestinal immune homeostasis by promoting mucosal immunity and limiting chronic inflammation.     

Classification and Categorical Inputs with Treed Gaussian Process Models

Recognizing the successes of treed Gaussian process (TGP) models as an interpretable and thrifty model for nonparametric regression, we seek to extend the model to classification. Both treed models and Gaussian processes (GPs) have, separately, enjoyed great success in application to classification problems. An example of the former is Bayesian CART. In the latter, real-valued GP output may be utilized for classification via latent variables, which provide classification rules by means of a softmax function. We formulate a Bayesian model averaging scheme to combine these two models and describe a Monte Carlo method for sampling from the full posterior distribution with joint proposals for the tree topology and the GP parameters corresponding to latent variables at the leaves. We concentrate on efficient sampling of the latent variables, which is important to obtain good mixing in the expanded parameter space. The tree structure is particularly helpful for this task and also for developing an efficient scheme for handling categorical predictors, which commonly arise in classification problems. Our proposed classification TGP (CTGP) methodology is illustrated on a collection of synthetic and real data sets. We assess performance relative to existing methods and thereby show how CTGP is highly flexible, offers tractable inference, produces rules that are easy to interpret, and performs well out of sample.     

Cell-to-cell spread of HIV permits ongoing replication despite antiretroviral therapy

Latency and ongoing replication have both been proposed to explain the drug-insensitive human immunodeficiency virus (HIV) reservoir maintained during antiretroviral therapy. Here we explore a novel mechanism for ongoing HIV replication in the face of antiretroviral drugs. We propose a model whereby multiple infections per cell lead to reduced sensitivity to drugs without requiring drug-resistant mutations, and experimentally validate the model using multiple infections per cell by cell-free HIV in the presence of the drug tenofovir. We then examine the drug sensitivity of cell-to-cell spread of HIV, a mode of HIV transmission that can lead to multiple infection events per target cell. Infections originating from cell-free virus decrease strongly in the presence of antiretrovirals tenofovir and efavirenz whereas infections involving cell-to-cell spread are markedly less sensitive to the drugs. The reduction in sensitivity is sufficient to keep multiple rounds of infection from terminating in the presence of drugs. We examine replication from cell-to-cell spread in the presence of clinical drug concentrations using a stochastic infection model and find that replication is intermittent, without substantial accumulation of mutations. If cell-to-cell spread has the same properties in vivo, it may have adverse consequences for the immune system, lead to therapy failure in individuals with risk factors, and potentially contribute to viral persistence and hence be a barrier to curing HIV infection.     

Multi-domain conformational selection underlies pre-mRNA splicing regulation by U2AF

Many cellular functions involve multi-domain proteins, which are composed of structurally independent modules connected by flexible linkers. Although it is often well understood how a given domain recognizes a cognate oligonucleotide or peptide motif, the dynamic interaction of multiple domains in the recognition of these ligands remains to be characterized. Here we have studied the molecular mechanisms of the recognition of the 3'-splice-site-associated polypyrimidine tract RNA by the large subunit of the human U2 snRNP auxiliary factor (U2AF65) as a key early step in pre-mRNA splicing. We show that the tandem RNA recognition motif domains of U2AF65 adopt two remarkably distinct domain arrangements in the absence or presence of a strong (that is, high affinity) polypyrimidine tract. Recognition of sequence variations in the polypyrimidine tract RNA involves a population shift between these closed and open conformations. The equilibrium between the two conformations functions as a molecular rheostat that quantitatively correlates the natural variations in polypyrimidine tract nucleotide composition, length and functional strength to the efficiency to recruit U2 snRNP to the intron during spliceosome assembly. Mutations that shift the conformational equilibrium without directly affecting RNA binding modulate splicing activity accordingly. Similar mechanisms of cooperative multi-domain conformational selection may operate more generally in the recognition of degenerate nucleotide or amino acid motifs by multi-domain proteins.     

Genome sequence and analysis of the tuber crop potato

Potato (Solanum tuberosum L.) is the world's most important non-grain food crop and is central to global food security. It is clonally propagated, highly heterozygous, autotetraploid, and suffers acute inbreeding depression. Here we use a homozygous doubled-monoploid potato clone to sequence and assemble 86% of the 844-megabase genome. We predict 39,031 protein-coding genes and present evidence for at least two genome duplication events indicative of a palaeopolyploid origin. As the first genome sequence of an asterid, the potato genome reveals 2,642 genes specific to this large angiosperm clade. We also sequenced a heterozygous diploid clone and show that gene presence/absence variants and other potentially deleterious mutations occur frequently and are a likely cause of inbreeding depression. Gene family expansion, tissue-specific expression and recruitment of genes to new pathways contributed to the evolution of tuber development. The potato genome sequence provides a platform for genetic improvement of this vital crop.