Improvements in viral gene annotation using large language models and soft alignments.

BACKGROUND: The annotation of protein sequences in public databases has long posed a challenge in molecular biology. This issue is particularly acute for viral proteins, which demonstrate limited homology to known proteins when using alignment, k-mer, or profile-based homology search approaches. A novel methodology employing Large Language Models (LLMs) addresses this methodological challenge by annotating protein sequences based on embeddings. RESULTS: Central to our contribution is the soft alignment algorithm, drawing from traditional protein alignment but leveraging embedding similarity at the amino acid level to bypass the need for conventional scoring matrices. This method not only surpasses pooled embedding-based models in efficiency but also in interpretability, enabling users to easily trace homologous amino acids and delve deeper into the alignments. Far from being a black box, our approach provides transparent, BLAST-like alignment visualizations, combining traditional biological research with AI advancements to elevate protein annotation through embedding-based analysis while ensuring interpretability. Tests using the Virus Orthologous Groups and ViralZone protein databases indicated that the novel soft alignment approach recognized and annotated sequences that both blastp and pooling-based methods, which are commonly used for sequence annotation, failed to detect. CONCLUSION: The embeddings approach shows the great potential of LLMs for enhancing protein sequence annotation, especially in viral genomics. These findings present a promising avenue for more efficient and accurate protein function inference in molecular biology.

Reversible Control of the Mn Oxidation State in SrTiO3 Bulk Powders.

We demonstrate a low-temperature reduction method for exhibiting fine control over the oxidation state of substitutional Mn ions in strontium titanate (SrTiO3) bulk powder. We employ NaBH4 as the chemical reductant that causes significant changes in the oxidation state and oxygen vacancy complexation with Mn2+ dopants at temperatures <350°C where lattice reduction is negligible. At higher reduction temperatures, we also observe the formation of Ti3+ in the lattice by diffuse-reflectance and low-temperature electron paramagnetic resonance (EPR) spectroscopy. In addition to Mn2+, Mn4+, and the Mn2+ complex with an oxygen vacancy, we also observe a sharp resonance in the EPR spectrum of heavily reduced Mn-doped SrTiO3. This sharp signal is tentatively assigned to surface superoxide ion that is formed by the surface electron transfer reaction between Ti3+ and O2. The ability to control the relative amounts of various paramagnetic defects in SrTiO3 provides many possibilities to study in a model system the impact of tunable dopant-defect interactions for spin-based electronic applications or visible-light photocatalysis.

Transferable Memristive Nanoribbons Comprising Solution-Processed Strontium Titanate Nanocubes.

Memristors, often comprising an insulating metal oxide film between two metal electrodes (MIM), constitute a class of two-terminal devices that possesses tunable variations in resistance based on the applied bias history. Intense research remains focused on the metal-insulator interface, which serves as the crux of coupled electronic-ionic interactions and dictates the underpinning transport mechanisms at either electrode. Top-down, ultrahigh-vacuum (UVH) deposition approaches for MIM nanostructures yield highly crystalline, heteroepitaxial interfaces but limit the number of electrode configurations due to a fixed bottom electrode. Here we report on the convective self-assembly, removal, and transfer of individual nanoribbons comprising solution-processed, single-crystalline strontium titanate (STO) perovskite oxide nanocrystals to arbitrary metallized substrates. Nanoribbon transferability enables changes in transport models ranging from interfacial trap-detrap to electrochemical metallization processes. We also demonstrate the endurance of memristive behavior, including switching ratios up to 104, after nanoribbon redeposition onto poly(ethylene terephthalate) (PET) flexible substrates. The combination of ambient, aerobic prepared nanocrystals and convective self-assembly deposition herein provides a pathway for facile, scalable manufacturing of high quality, functional oxide nanostructures on arbitrary surfaces and topologies.