Sparse transformer with local and seasonal adaptation for multivariate time series forecasting.

Transformers have achieved remarkable performance in multivariate time series(MTS) forecasting due to their capability to capture long-term dependencies. However, the canonical attention mechanism has two key limitations: (1) its quadratic time complexity limits the sequence length, and (2) it generates future values from the entire historical sequence. To address this, we propose a Dozer Attention mechanism consisting of three sparse components: (1) Local, each query exclusively attends to keys within a localized window of neighboring time steps.

A novel extreme adaptive GRU for multivariate time series forecasting.

Multivariate time series forecasting is a critical problem in many real-world scenarios. Recent advances in deep learning have significantly enhanced the ability to tackle such problems. However, a primary challenge in time series forecasting comes from the imbalanced time series data that include extreme events.

Bacterial susceptibility and resistance to modelin-5.

Modelin-5 (M5-NH) killed with a minimum lethal concentration (MLC) of 5.86 μM and strongly bound its cytoplasmic membrane (CM) with a of 23.5 μM.

A robust and accurate single-cell data trajectory inference method using ensemble pseudotime.

Background: The advance in single-cell RNA sequencing technology has enhanced the analysis of cell development by profiling heterogeneous cells in individual cell resolution. In recent years, many trajectory inference methods have been developed. They have focused on using the graph method to infer the trajectory using single-cell data, and then calculate the geodesic distance as the pseudotime.

PEGylation enhances the antibacterial and therapeutic potential of amphibian host defence peptides.

Aurein 2.1, aurein 2.6 and aurein 3.

Antimicrobial Peptides with pH-Dependent Activity and Alkaline Optima: Their Origins, Mechanisms of Action and Potential Applications.

A number of disorders and diseases are associated with conditions of high pH, and many conventional antibiotics lose their efficacy under these pH conditions, generating a need for novel antimicrobials. A potential solution to fulfill this need is Antimicrobial Peptides (AMPs) with high pH optima. This review shows that a variety of anionic and cationic AMPs with this pH dependency are produced by creatures across the eukaryotic kingdom, including rabbits, cattle, sheep, fish, crabs and frog.

Linearized esculentin-2EM shows pH dependent antibacterial activity with an alkaline optimum.

Here the hypothesis that linearized esculentin 2EM (E2EM-lin) from Glandirana emeljanovi possesses pH dependent activity is investigated. The peptide showed weak activity against Gram-negative bacteria (MLCs ≥ 75.0 μM) but potent efficacy towards Gram-positive bacteria (MLCs ≤ 6.

Biophysical studies on the antimicrobial activity of linearized esculentin 2EM.

Linearized esculentin 2 EM (E2EM-lin) from the frog, Glandirana emeljanovi was highly active against Gram-positive bacteria (minimum lethal concentration ≤ 5.0 μM) and strongly α-helical in the presence of lipid mimics of their membranes (>55.0%).

Biophysical investigation into the antibacterial action of modelin-5-NH.

Modelin-5-CONH2 (M5-NH2) is a synthetic antimicrobial peptide, which was found to show potent activity against Bacillus subtilis (minimum lethal concentration = 8.47 μM) and to bind strongly to membranes of the organism (Kd = 10.44 μM).

An Atlas of Anionic Antimicrobial Peptides from Amphibians.

Anionic antimicrobial peptides (AAMPs) with net charges ranging from -1 to -8 have been identified in frogs, toads, newts and salamanders across Africa, South America and China. Most of these peptides show antibacterial activity and a number of them are multifunctional, variously showing antifungal activity, anticancer action, neuropeptide function and the ability to potentiate conventional antibiotics. Antimicrobial mechanisms proposed for these AAMPs, include toroidal pore formation and the Shai-Huang-Matsazuki model of membrane interaction along with pH dependent amyloidogenesis and membranolysis via tilted peptide formation.

Investigations into the potential anticancer activity of Maximin H5.

Here we report the first major example of anionic amphibian host defence peptides (HDPs) with anticancer activity. Maximin H5 (MH5N) is a C-terminally amidated, anionic host defence peptide from toads of the Bombina genus, which was shown to possess activity against the glioma cell line, T98G (EC = 125 μM). The peptide adopted high levels of α-helical structure (57.

pH Dependent Antimicrobial Peptides and Proteins, Their Mechanisms of Action and Potential as Therapeutic Agents.

Antimicrobial peptides (AMPs) are potent antibiotics of the innate immune system that have been extensively investigated as a potential solution to the global problem of infectious diseases caused by pathogenic microbes. A group of AMPs that are increasingly being reported are those that utilise pH dependent antimicrobial mechanisms, and here we review research into this area. This review shows that these antimicrobial molecules are produced by a diverse spectrum of creatures, including vertebrates and invertebrates, and are primarily cationic, although a number of anionic examples are known.

Low pH Enhances the Action of Maximin H5 against Staphylococcus aureus and Helps Mediate Lysylated Phosphatidylglycerol-Induced Resistance.

Maximin H5 (MH5) is an amphibian antimicrobial peptide specifically targeting Staphylococcus aureus. At pH 6, the peptide showed an improved ability to penetrate (ΔΠ = 6.2 mN m(-1)) and lyse (lysis = 48%) Staphylococcus aureus membrane mimics, which incorporated physiological levels of lysylated phosphatidylglycerol (Lys-PG, 60%), compared to that at pH 7 (ΔΠ = 5.

Anionic Host Defence Peptides from the Plant Kingdom: Their Anticancer Activity and Mechanisms of Action.

It is becoming increasingly clear that plants ranging across the plant kingdom produce anionic host defence peptides (AHDPs) with potent activity against a wide variety of human cancers cells. In general, this activity involves membrane partitioning by AHDPs, which leads to membranolysis and / or internalization to attack intracellular targets such as DNA. Several models have been proposed to describe these events including: the toroidal pore and Shai-Matsuzaki-Huang mechanisms but, in general, the mechanisms underpinning the membrane interactions and anticancer activity of these peptides are poorly understood.

ModFossa: A library for modeling ion channels using Python.

The creation and simulation of ion channel models using continuous-time Markov processes is a powerful and well-used tool in the field of electrophysiology and ion channel research. While several software packages exist for the purpose of ion channel modeling, most are GUI based, and none are available as a Python library. In an attempt to provide an easy-to-use, yet powerful Markov model-based ion channel simulator, we have developed ModFossa, a Python library supporting easy model creation and stimulus definition, complete with a fast numerical solver, and attractive vector graphics plotting.

The increasing role of phosphatidylethanolamine as a lipid receptor in the action of host defence peptides.

Host defence peptides (HDPs) are antimicrobial agents produced by organisms across the prokaryotic and eukaryotic kingdoms. Many prokaryotes produce HDPs, which utilise lipid and protein receptors in the membranes of bacterial competitors to facilitate their antibacterial action and thereby survive in their niche environment. As a major example, it is well established that cinnamycin and duramycins from Streptomyces have a high affinity for phosphatidylethanolamine (PE) and exhibit activity against other Gram-positive organisms, such as Bacillus.

The role of C-terminal amidation in the membrane interactions of the anionic antimicrobial peptide, maximin H5.

Maximin H5 is an anionic antimicrobial peptide from amphibians, which carries a C-terminal amide moiety, and was found to be moderately haemolytic (20%). The α-helicity of the peptide was 42% in the presence of lipid mimics of erythrocyte membranes and was found able to penetrate (10.8 mN m(-1)) and lyse these model membranes (64 %).

Cn-AMP2 from green coconut water is an anionic anticancer peptide.

Globally, death due to cancers is likely to rise to over 20 million by 2030, which has created an urgent need for novel approaches to anticancer therapies such as the development of host defence peptides. Cn-AMP2 (TESYFVFSVGM), an anionic host defence peptide from green coconut water of the plant Cocos nucifera, showed anti-proliferative activity against the 1321N1 and U87MG human glioma cell lines with IC50 values of 1.25 and 1.

Sounding the death knell for microbes?

Over the past 5 years, several studies showed that ultrasound, which is sound with a frequency>20 kHz, is able to kill bacteria by activating molecules termed sonosensitizers (SS) to produce reactive oxygen species, which are toxic to microbes. It is our opinion that this work opens up the potential for the development of a novel form of ultrasound-mediated antimicrobial therapy. Termed sonodynamic antimicrobial chemotherapy (SACT), we define this therapy as a regime where a SS is selectively delivered to target microbial cells and activated by ultrasound to induce the death of those microbial cells.

Using sound for microbial eradication--light at the end of the tunnel?

Sonodynamic antimicrobial chemotherapy (SACT) is a novel modality, which uses ultrasound to kill bacteria by the activation of molecules termed sonosensitisers (SS) to produce reactive oxygen species that are toxic to microorganism although microbial resistance to this modality has been reported. There are a growing number of SS being reported with the dual ability to be activated by both ultrasound and light, and we hypothesis that a novel antimicrobial strategy, potentially known as sonophotodynamic antimicrobial chemotherapy (SPACT), could be developed based on these agents. SPACT offers advantages over SACT and could constitute a new weapon in the fight against the growing global threat posed by microbial infections.

The interaction of aurein 2.5 with fungal membranes.

Aurein 2.5 (GLFDIVKKVVGAFGSL-NH2) is an antimicrobial peptide, which was seen to have activity against Stachybotris chartarum, Penicillium roseopurpureum and Aspergillus flavus with minimum fungicidal concentrations in the range 250-500 μM. S.

Small-molecule clinical trial candidates for the treatment of glioma.

The evaluation of new glioma treatments, currently in clinical trials, is undertaken.

A novel CPU/GPU simulation environment for large-scale biologically realistic neural modeling.

Computational Neuroscience is an emerging field that provides unique opportunities to study complex brain structures through realistic neural simulations. However, as biological details are added to models, the execution time for the simulation becomes longer. Graphics Processing Units (GPUs) are now being utilized to accelerate simulations due to their ability to perform computations in parallel.

A novel form of bacterial resistance to the action of eukaryotic host defense peptides, the use of a lipid receptor.

Host defense peptides show great potential for development as new antimicrobial agents with novel mechanisms of action. However, a small number of resistance mechanisms to their action are known, and here, we report a novel bacterial resistance mechanism mediated by a lipid receptor. Maximin H5 from Bombina maxima bound anionic and zwitterionic membranes with low affinity (Kd > 225 μM) while showing a strong ability to lyse (>55%) and penetrate (π > 6.