Small open reading frames (sORFs; <300 nucleotides or <100 amino acids) are widespread across all genomes, and an increasing variety of them appear to be translating from non-genic regions. Over the past few decades, peptides produced from sORFs have been identified as functional in various organisms, from bacteria to humans. Despite recent advances in next-generation sequencing and proteomics, accurate annotation and classification of sORFs remain a rate-limiting step toward reliable and high-throughput detection of small proteins from non-genic regions. Additionally, the cost of computational methods utilizing machine learning is lower than that of biological experiments, and they can be employed to detect sORFs, laying the groundwork for biological experiments. We present D-sORF, a machine-learning framework that integrates the statistical nucleotide context and motif information around the start codon to predict coding sORFs. D-sORF scores directly for coding identity and requires only the underlying genomic sequence, without incorporating parameters such as the conservation, which, in the case of sORFs, may increase the dispersion of scores within the significantly less conserved non-genic regions. D-sORF achieves 94.74% precision and 92.37% accuracy for small ORFs (using the 99 nt medium length window). When D-sORF is applied to sORFs associated with ribosomes, the identification of transcripts producing peptides (annotated by the Ensembl IDs) is similar to or superior to experimental methodologies based on ribosome-sequencing (Ribo-Seq) profiling. In parallel, the recognition of putative negative data, such as the intron-containing transcripts that associate with ribosomes, remains remarkably low, indicating that D-sORF could be efficiently applied to filter out false-positive sORFs from Ribo-Seq data because of the non-productive ribosomal binding or noise inherent in these protocols.

The exponential growth of publicly available omics datasets and biomedical literature has created both opportunities and challenges for data-driven discovery in life sciences. While the Gene Expression Omnibus (GEO) hosts millions of high-throughput experimental datasets, the European Nucleotide Archive (ENA) stores the corresponding raw sequencing data, and PubMed contains an extensive body of related scientific publications, integrated exploration of these resources remains limited. We present Public Omics Explorer (POE), a web-based platform that performs literature‑informed dataset retrieval, semantically linking GEO datasets and ENA records through their associated PubMed publications. POE automatically collects and indexes GEO metadata, ENA cross‑references, and PubMed abstracts on a daily basis. For semantic embedding, POE employs the biomedical-specialized SBioBERT model, which generates dense vector representations from publication text. These embeddings are indexed using Facebook AI Similarity Search (FAISS) to enable high-precision, context-aware retrieval. Users can search using free‑text natural language queries, which are processed through semantic search to identify conceptually relevant datasets based on linked publication content. Structured filters allow refinement by organism, experiment type, library strategy, sample type, extracted molecule, and publication year. In addition to semantic queries, POE supports direct retrieval of datasets via accession identifiers (GSE IDs, PubMed IDs, DOIs) and offers a programmatic RESTful API for integration into computational pipelines and automated workflows. By linking processed data in GEO with raw data in ENA through shared publication context, POE facilitates hypothesis generation, meta‑analysis, and exploratory research. The application is freely available at https://nplab.gr/poe.