Cross-SEAN: A cross-stitch semi-supervised neural attention model for COVID-19 fake news detection.

As the COVID-19 pandemic sweeps across the world, it has been accompanied by a tsunami of fake news and misinformation on social media. At the time when reliable information is vital for public health and safety, COVID-19 related fake news has been spreading even faster than the facts. During times such as the COVID-19 pandemic, fake news can not only cause intellectual confusion but can also place people's lives at risk. This calls for an immediate need to contain the spread of such misinformation on social media. We introduce CTF, a large-scale COVID-19 Twitter dataset with labelled genuine and fake tweets. Additionally, we propose Cross-SEAN, a cross-stitch based semi-supervised end-to-end neural attention model which leverages the large amount of unlabelled data. Cross-SEAN partially generalises to emerging fake news as it learns from relevant external knowledge. We compare Cross-SEAN with seven state-of-the-art fake news detection methods. We observe that it achieves 0.95 F1 Score on CTF, outperforming the best baseline by 9%. We also develop Chrome-SEAN, a Cross-SEAN based chrome extension for real-time detection of fake tweets.

The Effect of Oligomerization on A Solid-Binding Peptide Binding to Silica-Based Materials.

The bifunctional linker-protein G (LPG) fusion protein comprises a peptide (linker) sequence and a truncated form of Streptococcus strain G148 protein G (protein G). The linker represents a multimeric solid-binding peptide (SBP) comprising 4 × 21-amino acid sequence repeats that display high binding affinity towards silica-based materials. In this study, several truncated derivatives were investigated to determine the effect of the SBP oligomerization on the silica binding function of LPG (for the sake of clarity, LPG will be referred from here on as 4 × LPG). Various biophysical characterization techniques were used to quantify and compare the truncated derivatives against 4 × LPG and protein G without linker (PG). The derivative containing two sequence repeats (2 × LPG) showed minimal binding to silica, while the truncated derivative with only a single sequence (1 × LPG) displayed no binding. The derivative containing three sequence repeats (3 × LPG) was able to bind to silica with a binding affinity of KD = 53.23 ± 4.5 nM, which is 1.5 times lower than that obtained for 4 × LPG under similar experimental conditions. Circular dichroism (CD) spectroscopy and fluorescence spectroscopy studies indicated that the SBP degree of oligomerization has only a small effect on the secondary structure (the linker unravels the beginning of the protein G sequence) and chemical stability of the parent protein G. However, based on quartz crystal microbalance with dissipation monitoring (QCM-D), oligomerization is an important parameter for a strong and stable binding to silica. The replacement of three sequence repeats by a (GGGGS)12 glycine-rich spacer indicated that the overall length rather than the SBP oligomerization mediated the effective binding to silica.

Elucidating the Binding Mechanism of a Novel Silica-Binding Peptide.

Linker-protein G (LPG) is a bifunctional fusion protein composed of a solid-binding peptide (SBP, referred as the "linker") with high affinity to silica-based compounds and a Streptococcus protein G (PG), which binds antibodies. The binding mechanisms of LPG to silica-based materials was studied using different biophysical techniques and compared to that of PG without the linker. LPG displayed high binding affinity to a silica surface (KD = 34.77 ± 11.8 nM), with a vertical orientation, in comparison to parent PG, which exhibited no measurable binding affinity. Incorporation of the linker in the fusion protein, LPG, had no effect on the antibody-binding function of PG, which retained its secondary structure and displayed no alteration of its chemical stability. The LPG system provided a milder, easier, and faster affinity-driven immobilization of antibodies to inorganic surfaces when compared to traditional chemical coupling techniques.

Experimental and theoretical tools to elucidate the binding mechanisms of solid-binding peptides.

The interactions between biomolecules and solid surfaces play an important role in designing new materials and applications which mimic nature. Recently, solid-binding peptides (SBPs) have emerged as potential molecular building blocks in nanobiotechnology. SBPs exhibit high selectivity and binding affinity towards a wide range of inorganic and organic materials. Although these peptides have been widely used in various applications, there is a need to understand the interaction mechanism between the peptide and its material substrate, which is challenging both experimentally and theoretically. This review describes the main characterisation techniques currently available to study SBP-surface interactions and their contribution to gain a better insight for designing new peptides for tailored binding.