Molecular profiling of thyroid cancer subtypes using large-scale text mining

Background

Thyroid cancer is the most common endocrine tumor with a steady increase in incidence. It is classified into multiple histopathological subtypes with potentially distinct molecular mechanisms. Identifying the most relevant genes and biological pathways reported in the thyroid cancer literature is vital for understanding of the disease and developing targeted therapeutics.

Results

We developed a large-scale text mining system to generate a molecular profiling of thyroid cancer subtypes. The system first uses a subtype classification method for the thyroid cancer literature, which employs a scoring scheme to assign different subtypes to articles. We evaluated the classification method on a gold standard derived from the PubMed Supplementary Concept annotations, achieving a micro-average F1-score of 85.9% for primary subtypes. We then used the subtype classification results to extract genes and pathways associated with different thyroid cancer subtypes and successfully unveiled important genes and pathways, including some instances that are missing from current manually annotated databases or most recent review articles.

Conclusions

Identification of key genes and pathways plays a central role in understanding the molecular biology of thyroid cancer. An integration of subtype context can allow prioritized screening for diagnostic biomarkers and novel molecular targeted therapeutics. Source code used for this study is made freely available online at https://github.com/chengkun-wu/GenesThyCan.

     

GABAA Receptors Containing ρ1 Subunits Contribute to In Vivo Effects of Ethanol in Mice

GABA_A receptors consisting of ρ1, ρ2, or ρ3 subunits in homo- or hetero-pentamers have been studied mainly in retina but are detected in many brain regions. Receptors formed from ρ1 are inhibited by low ethanol concentrations, and family-based association analyses have linked ρ subunit genes with alcohol dependence. We determined if genetic deletion of ρ1 in mice altered in vivo ethanol effects. Null mutant male mice showed reduced ethanol consumption and preference in a two-bottle choice test with no differences in preference for saccharin or quinine. Null mutant mice of both sexes demonstrated longer duration of ethanol-induced loss of righting reflex (LORR), and males were more sensitive to ethanol-induced motor sedation. In contrast, ρ1 null mice showed faster recovery from acute motor incoordination produced by ethanol. Null mutant females were less sensitive to ethanol-induced development of conditioned taste aversion. Measurement of mRNA levels in cerebellum showed that deletion of ρ1 did not change expression of ρ2, α2, or α6 GABA_A receptor subunits. (S)-4-amino-cyclopent-1-enyl butylphosphinic acid (“ρ1” antagonist), when administered to wild type mice, mimicked the changes that ethanol induced in ρ1 null mice (LORR and rotarod tests), but the ρ1 antagonist did not produce these effects in ρ1 null mice. In contrast, (R)-4-amino-cyclopent-1-enyl butylphosphinic acid (“ρ2” antagonist) did not change ethanol actions in wild type but produced effects in mice lacking ρ1 that were opposite of the effects of deleting (or inhibiting) ρ1. These results suggest that ρ1 has a predominant role in two in vivo effects of ethanol, and a role for ρ2 may be revealed when ρ1 is deleted. We also found that ethanol produces similar inhibition of function of recombinant ρ1 and ρ2 receptors. These data indicate that ethanol action on GABA_A receptors containing ρ1/ρ2 subunits may be important for specific effects of ethanol in vivo.     

Baseline Natural Killer and T Cell Populations Correlation with Virologic Outcome after Regimen Simplification to Atazanavir/Ritonavir Alone (ACTG 5201)

Objectives

Simplified maintenance therapy with ritonavir-boosted atazanavir (ATV/r) provides an alternative treatment option for HIV-1 infection that spares nucleoside analogs (NRTI) for future use and decreased toxicity. We hypothesized that the level of immune activation (IA) and recovery of lymphocyte populations could influence virologic outcomes after regimen simplification.

Methods

Thirty-four participants with virologic suppression ≥48 weeks on antiretroviral therapy (2 NRTI plus protease inhibitor) were switched to ATV/r alone in the context of the ACTG 5201 clinical trial. Flow cytometric analyses were performed on PBMC isolated from 25 patients with available samples, of which 24 had lymphocyte recovery sufficient for this study. Assessments included enumeration of T-cells (CD4/CD8), natural killer (NK) (CD3⁺CD56⁺CD16⁺) cells and cell-associated markers (HLA-DR, CD''s 38/69/94/95/158/279).

Results

Eight of the 24 patients had at least one plasma HIV-1 RNA level (VL) >50 copies/mL during the study. NK cell levels below the group median of 7.1% at study entry were associated with development of VL >50 copies/mL following simplification by regression and survival analyses (p = 0.043 and 0.023), with an odds ratio of 10.3 (95% CI: 1.92–55.3). Simplification was associated with transient increases in naïve and CD25⁺ CD4⁺ T-cells, and had no impact on IA levels.

Conclusions

Lower NK cell levels prior to regimen simplification were predictive of virologic rebound after discontinuation of nucleoside analogs. Regimen simplification did not have a sustained impact on markers of IA or T lymphocyte populations in 48 weeks of clinical monitoring.

Trial Registration

ClinicalTrials.gov {"type":"clinical-trial","attrs":{"text":"NCT00084019","term_id":"NCT00084019"}}NCT00084019     

Neither the HIV Protease Inhibitor Lopinavir-Ritonavir nor the Antimicrobial Trimethoprim-Sulfamethoxazole Prevent Malaria Relapse in Plasmodium cynomolgi-Infected Non-Human Primates

Plasmodium vivax malaria causes significant morbidity and mortality worldwide, and only one drug is in clinical use that can kill the hypnozoites that cause P. vivax relapses. HIV and P. vivax malaria geographically overlap in many areas of the world, including South America and Asia. Despite the increasing body of knowledge regarding HIV protease inhibitors (HIV PIs) on P. falciparum malaria, there are no data regarding the effects of these treatments on P. vivax''s hypnozoite form and clinical relapses of malaria. We have previously shown that the HIV protease inhibitor lopinavir-ritonavir (LPV-RTV) and the antibiotic trimethoprim sulfamethoxazole (TMP-SMX) inhibit Plasmodium actively dividing liver stages in rodent malarias and in vitro in P. falciparum, but effect against Plasmodium dormant hypnozoite forms remains untested. Separately, although other antifolates have been tested against hypnozoites, the antibiotic trimethoprim sulfamethoxazole, commonly used in HIV infection and exposure management, has not been evaluated for hypnozoite-killing activity. Since Plasmodium cynomolgi is an established animal model for the study of liver stages of malaria as a surrogate for P. vivax infection, we investigated the antimalarial activity of these drugs on Plasmodium cynomolgi relapsing malaria in rhesus macaques. Herein, we demonstrate that neither TMP-SMX nor LPV-RTV kills hypnozoite parasite liver stage forms at the doses tested. Because HIV and malaria geographically overlap, and more patients are being managed for HIV infection and exposure, understanding HIV drug impact on malaria infection is important.     

Availability of cadmium and zinc accumulated in the leaves of Thlaspi caerulescens incorporated into soil

When grown on contaminated soil, hyperaccumulator plants contain high concentrations of metals which may return to the soil after senescence. This work was undertaken to assess the availability of Cd and Zn associated to the leaves of the hyperaccumulator Thlaspi caerulescens after incorporation into an uncontaminated soil. A Zn- and Cd- accumulator population of T. caerulescens was grown on a Cd- and Zn- contaminated soil previously labelled with ¹⁰⁹Cd. Leaves (TCL) were harvested, dried, ground and incorporated into the soil at a rate of 2.07 mg Cd kg⁻¹ and 51.9 mg Zn kg⁻¹. Then a pot experiment was conducted for 3 months with rye grass (Lolium perenne) and T. caerulescens. Rye grass was harvested monthly and T. caerulescens at the end of the experiment. Plant biomass was measured, along with the concentration of Cd, Zn and ¹⁰⁹Cd. Results showed that water-extractable metals in TCL were 69% for Zn and 33% for Cd. Addition of TCL to soil, depleted growth of rye grass, and improved that of T. caerulescens. At harvest, concentrations of both metals were increased in plants by TCL. Concentrations of Cd in rye grass increased with the cut number, while that of Zn decreased slightly. Rye grass extracted 1.6% of the total Cd and 0.9% of the total Zn, and T. caerulescens extracted up to 22.4% of the Cd and 7% of the Zn. About 94% of the Cd in rye grass and 86% in T. caerulescens was derived from TCL. In conclusion, metals associated with leaves of the hyperaccumulator T. caerulescens were very mobile after incorporation into the soil. This revised version was published online in June 2006 with corrections to the Cover Date.     

A greedy algorithm for aligning DNA sequences.

For aligning DNA sequences that differ only by sequencing errors, or by equivalent errors from other sources, a greedy algorithm can be much faster than traditional dynamic programming approaches and yet produce an alignment that is guaranteed to be theoretically optimal. We introduce a new greedy alignment algorithm with particularly good performance and show that it computes the same alignment as does a certain dynamic programming algorithm, while executing over 10 times faster on appropriate data. An implementation of this algorithm is currently used in a program that assembles the UniGene database at the National Center for Biotechnology Information.     

Understanding Actin Organization in Cell Structure through Lattice Based Monte Carlo Simulations

Understanding the connection between mechanics and cell structure requires the exploration of the key molecular constituents responsible for cell shape and motility. One of these molecular bridges is the cytoskeleton, which is involved with intracellular organization and mechanotransduction. In order to examine the structure in cells, we have developed a computational technique that is able to probe the self-assembly of actin filaments through a lattice based Monte Carlo method. We have modeled the polymerization of these filaments based upon the interactions of globular actin through a probabilistic model encompassing both inert and active proteins. The results show similar response to classic ordinary differential equations at low molecular concentrations, but a bi-phasic divergence at realistic concentrations for living mammalian cells. Further, by introducing localized mobility parameters, we are able to simulate molecular gradients that are observed in non-homogeneous protein distributionsin vivo. The method and results have potential applications in cell and molecular biology as well as self-assembly for organic and inorganic systems.