From tradition to innovation: conventional and deep learning frameworks in genome annotation.

Following the milestone success of the Human Genome Project, the 'Encyclopedia of DNA Elements (ENCODE)' initiative was launched in 2003 to unearth information about the numerous functional elements within the genome. This endeavor coincided with the emergence of numerous novel technologies, accompanied by the provision of vast amounts of whole-genome sequences, high-throughput data such as ChIP-Seq and RNA-Seq. Extracting biologically meaningful information from this massive dataset has become a critical aspect of many recent studies, particularly in annotating and predicting the functions of unknown genes. The core idea behind genome annotation is to identify genes and various functional elements within the genome sequence and infer their biological functions. Traditional wet-lab experimental methods still rely on extensive efforts for functional verification. However, early bioinformatics algorithms and software primarily employed shallow learning techniques; thus, the ability to characterize data and features learning was limited. With the widespread adoption of RNA-Seq technology, scientists from the biological community began to harness the potential of machine learning and deep learning approaches for gene structure prediction and functional annotation. In this context, we reviewed both conventional methods and contemporary deep learning frameworks, and highlighted novel perspectives on the challenges arising during annotation underscoring the dynamic nature of this evolving scientific landscape.

Triplet-Based Deep Hashing Network for Cross-Modal Retrieval.

Given the benefits of its low storage requirements and high retrieval efficiency, hashing has recently received increasing attention. In particular, cross-modal hashing has been widely and successfully used in multimedia similarity search applications. However, almost all existing methods employing cross-modal hashing cannot obtain powerful hash codes due to their ignoring the relative similarity between heterogeneous data that contains richer semantic information, leading to unsatisfactory retrieval performance. In this paper, we propose a tripletbased deep hashing (TDH) network for cross-modal retrieval. First, we utilize the triplet labels, which describes the relative relationships among three instances as supervision in order to capture more general semantic correlations between cross-modal instances. We then establish a loss function from the inter-modal view and the intra-modal view to boost the discriminative abilities of the hash codes. Finally, graph regularization is introduced into our proposed TDH method to preserve the original semantic similarity between hash codes in Hamming space. Experimental results show that our proposed method outperforms several state-of-the-art approaches on two popular cross-modal datasets.

Magnetic properties of conducting polymer nanostructures.

Magnetic susceptibility measurements on conducting polyaniline and polypyrrole nanostructures with different dopant type and doping level as functions of temperature and magnetic field are reported. The susceptibility data cannot be simply described as Curie-like susceptibility at lower temperatures and temperature-independent Pauli-like susceptibility at higher temperatures; some unusual transitions are observed in the temperature dependence of susceptibility, for example, paramagnetic susceptibility decreases gradually with lowering temperature, which suggests the coexistence of polarons and spinless bipolarons and possible formation of bipolarons with changing temperature or doping level. In particular, it is found that the direct current magnetic susceptibilities are strongly dependent on applied magnetic field, dopant type, and doping level.

Electromagnetic functionalized and core-shell micro/nanostructured polypyrrole composites.

Core-shell micro/nanostructured and electromagnetic functionalized polypyrrole (PPy) composites were prepared by a self-assembly process associated with the template method in the presence of p-toluenesulfonate acid (p-TSA) as the dopant, in which the spherical hydroxyl iron (Fe[OH], 0.5-5 microm in diameter) functioned not only as the hard template, but also as the "core" of the micro/nanostructure, and the self-assembled PPy-p-TSA nanofibers (20-30 nm in diameter) acted as the "shell" (50-100 nm in thickness) of the microspheres. We found that the core-shell micro/nanostructures exhibit controllable electromagnetic properties by adjusting the mass ratio of Fe[OH] to pyrrole monomer. The micelle model was proposed to interpret the self-assembly of the core-shell micro/nanostructured composites.

Preparation of highly conductive, self-assembled gold/polyaniline nanocables and polyaniline nanotubes.

One-dimensional gold/polyaniline (Au/PANI-CSA) coaxial nanocables with an average diameter of 50-60 nm and lengths of more than 1 mum were successfully synthesized by reacting aniline monomer with chlorauric acid (HAuCl(4)) through a self-assembly process in the presence of D-camphor-10-sulfonic acid (CSA), which acts as both a dopant and surfactant. It was found that the formation probability and the size of the Au/PANI-CSA nanocables depends on the molar ratio of aniline to HAuCl(4) and the concentration of CSA, respectively. A synergistic growth mechanism was proposed to interpret the formation of the Au/PANI-CSA nanocables. The directly measured conductivity of a single gold/polyaniline nanocable was found to be high (approximately 77.2 S cm(-1)). Hollow PANI-CSA nanotubes, with an average diameter of 50-60 nm, were also obtained successfully by dissolving the Au nanowire core of the Au/PANI-CSA nanocables.

Synthesis, property and field-emission behaviour of amorphous polypyrrole nanowires.

Polypyrrole nanowires have been electrosynthesized by direct oxidation of 0.1 mol l(-1) pyrrole in a medium of 75% isopropyl alcohol + 20% boron trifluoride diethyl etherate + 5% poly (ethylene glycol) (by volume) using porous alumina membranes as the templates. The as-prepared nanowires had a smooth surface and uniform diameter and were arranged in an orderly manner in a high density. The conductivity of a single nanowire was measured by the four-electrode technique to be 23.4 S cm(-1) at room temperature. The field emission devices based on the nanowire array were fabricated and their operations were explored. The experimental results indicated that the field emission characteristics of the devices fitted well to the Fowler-Nordheim model of emission. The turn-on electric field was only 1.2 V microm(-1) and the current density reached 200 microA cm(-2) at 2.6 V microm(-1).