Coupling SIMD and SIMT Architectures to Boost Performance of a Phylogeny-aware Alignment Kernel

ABSTRACT: BACKGROUND: Aligning short DNA reads to a reference sequence alignment is a prerequisite fordetecting their biological origin and analyzing them in a phylogenetic context. With thePaPaRa tool we introduced a dedicated dynamic programming algorithm forsimultaneously aligning short reads to reference alignments and correspondingevolutionary reference trees. The algorithm aligns short reads to phylogenetic profiles thatcorrespond to the branches of such a reference tree. The algorithm needs to perform animmense number of pairwise alignments. Therefore, we explore vector intrinsics andGPUs to accelerate the PaPaRa alignment kernel. RESULTS: We optimized and parallelized PaPaRa on CPUs and GPUs. Via SSE 4.1 SIMD (SingleInstruction, Multiple Data) intrinsics for x86 SIMD architectures and multi-threading, weobtained a 9-fold acceleration on a single core as well as linear speedups with respect tothe number of cores. The peak CPU performance amounts to 18.1 GCUPS (Giga CellUpdates per Second) using all four physical cores on an Intel i7 2600 CPU running at 3.4GHz. The average CPU performance (averaged over all test runs) is 12.33 GCUPS. Wealso used OpenCL to execute PaPaRa on a GPU SIMT (Single Instruction, MultipleThreads) architecture. A NVIDIA GeForce 560 GPU delivered peak and averageperformance of 22.1 and 18.4 GCUPS respectively. Finally, we combined the SIMD andSIMT implementations into a hybrid CPU-GPU system that achieved an accumulatedpeak performance of 33.8 GCUPS. CONCLUSIONS: This accelerated version of PaPaRa (available at www.exelixis-lab.org/software.html)provides a significant performance improvement that allows for analyzing larger datasetsin less time. We observe that state-of-the-art SIMD and SIMT architectures delivercomparable performance for this dynamic programming kernel when the "competingprogrammer approach" is deployed. Finally, we show that overall performance can besubstantially increased by designing a hybrid CPU-GPU system with appropriate loaddistribution mechanisms.     

Immunodeficiency and autoimmunity: lessons from systemic lupus erythematosus

Recent evidence suggests that systemic autoimmunity and immunodeficiency are not separate entities, but rather are interconnected processes. Immunodeficiency results from distinct defects of the immune response and primarily presents as infections but also frequently with autoimmune features. Systemic autoimmunity is the combined effect of multiple genetic variations and infectious and immunoregulatory factors that result in dominant autoimmune manifestations, in addition to frequent and opportunistic infections. The overlap in disease manifestations and symptoms suggests that immunodeficiency should be considered in the presence of autoimmunity, and vice versa. In this review, we present the shared or similar aspects of immunodeficiency and autoimmunity using systemic lupus erythematosus as a paradigm and discuss the implications for clinical care.     

Stepwise binding of tylosin and erythromycin to Escherichia coli ribosomes, characterized by kinetic and footprinting analysis

Erythromycin and tylosin are 14- and 16-membered lactone ring macrolides, respectively. The current work shows by means of kinetic and chemical footprinting analysis that both antibiotics bind to Escherichia coli ribosomes in a two-step process. The first step established rapidly, involves a low-affinity binding site placed at the entrance of the exit tunnel in the large ribosomal subunit, where macrolides bind primarily through their hydrophobic portions. Subsequently, slow conformational changes mediated by the antibiotic hydrophilic portion push the drugs deeper into the tunnel, in a high-affinity site. Compared with erythromycin, tylosin shifts to the high-affinity site more rapidly, due to the interaction of the mycinose sugar of the drug with the loop of H35 in domain II of 23 S rRNA. Consistently, mutations of nucleosides U2609 and U754 implicated in the high-affinity site reduce the shift of tylosin to this site and destabilize, respectively, the final drug-ribosome complex. The weak interaction between tylosin and the ribosome is Mg2+ independent, unlike the tight binding. In contrast, both interactions between erythromycin and the ribosome are reduced by increasing concentrations of Mg2+ ions. Polyamines attenuate erythromycin affinity for the ribosome at both sequential steps of binding. In contrast, polyamines facilitate the initial binding of tylosin, but exert a detrimental, more pronounced, effect on the drug accommodation at its final position. Our results emphasize the role of the particular interactions that side chains of tylosin and erythromycin establish with 23 S rRNA, which govern the exact binding process of each drug and its response to the ionic environment.     

Effect of synthetic peptides corresponding to residues 313-332 of the alphaIIb subunit on platelet activation and fibrinogen binding to alphaIIbbeta3.

The platelet integrin receptor alphaIIbbeta3 plays a critical role in thrombosis and haemostasis by mediating interactions between platelets and several ligands but primarily fibrinogen. It has been shown previously that the YMESRADR KLAEVGRVYLFL (313-332) sequence of the alphaIIb subunit plays an important role in platelet activation, fibrinogen binding and alphaIIbbeta3-mediated outside-in signalling. Furthermore, we recently showed that the 20-residue peptide (20-mer) alphaIIb 313-332, is a potent inhibitor of platelet aggregation and fibrinogen binding to alphaIIbbeta3, interacting with fibrinogen rather than the receptor. In an effort to determine the sequence and the minimum length required for the biological activity of the above 20-mer, we synthesized seven octapeptides, each overlapping by six residues, covering the entire sequence and studied their effect on platelet activation as well as fibrinogen binding to activated platelets. We show for the first time that octapeptides containing the RAD sequence are capable of inhibiting platelet aggregation and secretion as well as fibrinogen binding to the activated alphaIIbbeta3, possibly interacting with the ligand rather than the receptor. This suggests that the RAD sequence, common to all the inhibitory peptides, is critical for their biological activity. However, the presence of the YMES sequence, adjacent to RAD, significantly increases the peptide's biological potency. The development of such inhibitors derived from the 313-332 region of the alphaIIb subunit may be advantageous against the RGD-like antagonists as they could inhibit platelet activation without interacting with alphaIIbbeta3, thus failing to further induce alphaIIbbeta3-mediated outside-in signalling.     

Decorin-induced growth inhibition is overcome through protracted expression and activation of epidermal growth factor receptors in osteosarcoma cells

Decorin is an established natural oncosuppressive factor whose action is being studied in detail. Recently, decorin gene therapy formulations using adenoviral vectors have been shown in several animal models with very promising results. The present study describes the first exception to the established oncosuppression model using human osteosarcoma cells. MG-63 osteosarcoma cells were found to constitutively produce decorin, and furthermore, to be resistant to decorin-induced growth arrest. On the contrary, decorin seemed to be beneficial to osteosarcoma cells because it was necessary for MG-63 cell migration and acted as a mediator, counteracting the transforming growth factor-beta2-induced cytostatic function. Efforts to determine how MG-63 cells could overcome the decorin-induced cytostatic effect established that decorin in MG-63 cells does not induce p21 expression nor does it cause protracted retraction and inactivation of the epidermal growth factor receptor. Conversely, epidermal growth factor receptor seemed to be overexpressed and continuously phosphorylated. In view of the proposed design of decorin-based anticancer therapeutic strategies, our study provides new data on pathways that cancer cells might employ to overcome the established decorin-induced growth suppression.     

Does obesity play a major role in the pathogenesis of sleep apnoea and its associated manifestations via inflammation, visceral adiposity, and insulin resistance?

Despite the early recognition of the strong association between obstructive sleep apnoea (OSA) and obesity, and OSA and cardiovascular problems, sleep apnoea has been treated as a "local abnormality" of the respiratory track rather than as a "systemic illness". In 1997, we first reported that the pro-inflammatory cytokines interleukin-6 (IL-6) and tumour necrosis factor-alpha (TNFalpha) were elevated in patients with disorders of excessive daytime sleepiness (EDS) and proposed that these cytokines were mediators of daytime sleepiness. In subsequent studies, it was shown that IL-6, TNFalpha, and insulin levels were elevated in sleep apnoea independently of obesity and that visceral fat was the primary parameter linked with sleep apnoea. Further studies showed that women with the polycystic ovary syndrome (PCOS) were much more likely than controls to have sleep-disordered breathing (SDB) and daytime sleepiness, suggesting a pathogenetic role of insulin resistance in OSA. Additional accumulated evidence that supports the role of obesity and the associated metabolic aberrations in the pathogenesis of sleep apnoea and related symptoms include: obesity without sleep apnoea is associated with daytime sleepiness; the protective role of gonadal hormones as suggested by the increased prevalence of sleep apnoea in post-menopausal women and the significantly reduced risk for OSA in women on hormonal therapy; partial effects of continuous positive airway pressure (CPAP) in obese patients with apnoea on hypercytokinemia, insulin resistance indices, and visceral fat; and that the prevalence of the metabolic syndrome in the U.S. population from the Third National Health and Nutrition Examination Survey (1988-1994) parallels the prevalence of symptomatic sleep apnoea in general random samples. Furthermore, the beneficial effect of a cytokine antagonist on EDS and apnoea in obese, male apnoeics and that of exercise and weight loss on SDB and EDS in general random or clinical samples, supports the hypothesis that cytokines and insulin resistance are mediators of EDS and sleep apnoea in humans. Finally, our recent finding that in obese, hypothalamic CRH neuron is hypoactive, provides additional evidence on the potential central neural mechanisms for depressed ventilation and consequent development of sleep apnoea in obese individuals. In conclusion, accumulating evidence provides support to our thesis that obesity via inflammation, insulin resistance, visceral adiposity, and central neural mechanisms, e.g. hypofunctioning hypothalamic CRH, play a major role in the pathogenesis of sleep apnoea, sleepiness, and the associated cardiovascular co-morbidities.     

Enzyme-assisted synthesis and structure characterization of glucuronic acid conjugates of losartan, candesartan, and zolarsartan

Three angiotensin II receptor antagonists—losartan, candesartan, and zolarsartan—were investigated. All the compounds, which are structural analogues, are metabolized via conjugation to glucuronic acid. Interestingly, both O- and N-glucuronidation take place, so that regioisomers are formed. One ether O-glucuronide, two acyl O-glucuronides, and five tetrazole-N-glucuronides were biosynthesized, in milligram scale, from the three sartan aglycones. Liver microsomes from bovine, moose, rat, and pig and recombinant human UDP-glucuronosyltransferases were used as catalysts. The synthesized compounds were identified as sartan glucuronides by mass spectrometry, while the sites of glucuronidation were determined by nuclear magnetic resonance spectroscopy. Drug metabolites are needed as standards for pharmaceutical research and, as the present study shows, they can easily be produced with enzymes as catalyst.     

Processing of DNA damage clusters in human cells: current status of knowledge

Eukaryotic cells exposed to DNA damaging agents activate important defensive pathways by inducing multiple proteins involved in DNA repair, cell cycle checkpoint control and potentially apoptosis. After the acceptance of the hypothesis that oxidatively generated clustered DNA lesions (OCDL: closely spaced DNA lesions) can be induced even by low doses of ionizing radiation or even endogenously, and significant advances have been made in the understanding of the biochemistry underlying the repair of closely spaced DNA lesions, many questions still remain unanswered. The major questions that have to be answered in the near future are: 1) how human cells process these types of DNA damage if they repair them at all, 2) under what conditions a double strand break (DSB) may be created during the processing of two closely spaced DNA lesions and 3) what type of repair protein interactions govern the processing of complex DNA damage? The data existing so far on human cells and tissues are very limited and in some cases contradicting. All of them though agree however on the major importance of gaining mechanistic insights on the pathways used by the cell to confront and process complex DNA damage located in a small DNA volume and the need of more in depth analytical studies. We selectively review recently-obtained data on the processing of non-DSB DNA damage clusters in human cells and tissues and discuss the current status of knowledge in the field.     

Atypical femoral fracture in a metastatic bone disease patient six months after discontinuation of denosumab received sequentially to previous bisphosphonate therapy - A case report

Although, both bisphosphonates and denosumab are effective in reducing the risk of skeletal-related events in patients with metastatic bone disease, many concerns were being raised about the possible association between their use and atypical femoral fractures. A case of an atypical femoral fracture in a metastatic bone disease patient, six months after discontinuation of long-term zoledronic acid therapy and sequential treatment with denosumab is reported. After extensive laboratory and imaging examination, the fracture was classified as atypical and it was finally treated with discontinuation of denosumab, long cephalomedullary interlocking nailing and vitamin D administration. Sequential treatment with bisphosphonates and denosumab in patients with metastatic bone disease, may lead to an overlapping treatment effect, increasing bone suppression and the risk of atypical femoral fracture. In addition, discontinuation of denosumab may activate bone remodeling units in an area with microdamage accumulation in cortical bone caused by the previous bone suppression from the antiresorptive treatment. The activation of bone remodeling units may accelerate the occurrence of the atypical femoral fractures.