Python ecosystem dominates the deep learning framework. flodl-bio is a Rust crate that extends flodl, a libtorch-backed deep-learning framework, with deep learning on biological sequences. To our knowledge, flodl-bio is the first Rust crate that provides a comprehensive and practical deep learning for the biological applications, which includes the alphabet-aware sequence encoding, including one-hot and k-mer representations, padded sequences, and channels-first batches. It also includes several FASTA parsing and graph construction utilities for adjacency matrices, GCN-normalised adjacency matrices, graph Laplacians, residue contact maps, and de Bruijn graphs. Among the added functionalities, it includes AUROC and AUPRC with their associated curves; classification metrics, and the export of learned convolutional filters as MEME motifs and seeded splits. Deep biological learning includes a multi-width motif CNN, a dilated residual CNN, bidirectional LSTM and GRU architectures, and transformer encoder. On the five human transcription-factor-binding datasets in the GUE benchmark, a motif CNN with approximately 10,000 parameters, trained from scratch, outperforms all published CNN baselines. It also superseded the 2.5-billion-parameter Nucleotide Transformer on two datasets and DNABERT-2 on one. Given a fixed hardware configuration and software stack, experiments are reproducible at the bit level. See Supplementary Table S2 of Supplementary Material. Availability and implementation: flodl-bio is written in Rust and released under the MIT licence at https://github.com/sablokg/flodl-bio . The core builds with stable Rust and has no dependencies. The model tier is enabled with the flodl Cargo feature (cuda for NVIDIA GPUs) and requires libtorch and Rust 1.91 or later. The evaluation harness is in the repository’s experiments directory, and the complete run outputs are archived at https://doi.org/10.5281/zenodo.23321850 .

Machine-learning models used in bioinformatics are exposed to covariate shift from new sequencing platforms, batches and cohorts, while ground-truth labels arrive late or never. We present nannyrust, a pure-Rust library with no third-party dependencies that implements univariate drift statistics (Kolmogorov–Smirnov, Wasserstein, Jensen–Shannon, Hellinger, chi-squared, L-infinity), classification metrics with ROC AUC, and confidence-based performance estimation (CBPE), so that monitoring can run inside compiled pipelines. Against SciPy and scikit-learn the core statistics agree to within 5 × 10⁻¹³ on 700 test datasets.

Modernized deep learning includes sequence models, graph neural networks and domain specific analysis. We present biotorch, a PyTorch-based Python package that handles biological sequences (DNA, RNA, protein), biological networks (protein–protein interaction, gene regulatory, metabolic and other typed, weighted or directed networks) and graphs (protein structures, molecules, de Bruijn graphs) under one interface. Each and every model exposes batch method, which allows the users to automate classification, regression, per-residue labelling, masked language modelling, node classification, link prediction, graph classification and sequence-network fusion. The package include synthetic benchmarks. biotorch is written in Python (≥3.9) and distributed as an installable package. Source code, documentation and tests: https://github.com/sablokg/biotorch. License: MIT. The package was developed and tested with Python 3.12 and PyTorch 2.14 on CPU.

Plant microbiome research is a rapidly developing and productive field, with substantial attention directed toward the identification of the major determinants of plant growth, nutrient acquisition, stress tolerance, disease resistance, and associated agricultural productivity. Advances in high-throughput sequencing and multi-omics technologies have generated large and complex datasets describing plant-associated microbial communities. However, the high dimensionality, sparsity, compositionality, and heterogeneity of microbiome datasets create substantial analytical challenges for conventional statistical approaches. Machine learning and artificial intelligence (AI) provide new opportunities to identify microbial biomarkers, predict plant phenotypes, characterize plant–microbe interactions, and design beneficial microbial communities. Applications include plant disease prediction, crop productivity, abiotic-stress tolerance, nutrient cycling, biomarker discovery, microbiome-assisted breeding, and synthetic microbial community (SynCom) design. Recent studies increasingly combine AI with multi-omics, synthetic biology, and predictive modeling, indicating a transition from descriptive microbiome analysis toward microbiome engineering. Nevertheless, model generalizability remains constrained by small sample sizes, compositional data, technical variation, batch effects, environmental heterogeneity, and insufficient external validation. The combination of microbiome science and AI has considerable potential to support sustainable, climate-resilient crop production, but successful translation will require rigorous biological validation and field-scale testing.

Biological computing needs the application of the high throughput machine learning approach to biological sequences. DNA, RNA and protein are the building blocks of the biological sequences and to use them in biological deep learning several sequence based tokentization such as one-hot, k-mer encodings, character, k-mer and trainable byte-pair-encoding (BPE) tokenizers, BERT-style masked-language-model, FASTA/FASTQ input and output are needed. Additionally to use them in the biological deep learning, padding-aware Transformer encoders, dilated residual CNNs, bidirectional LSTMs, and a masked language model are needed. Bio-keras fills this gap with the ready to use the tokenizers and the bio-layers for machine and deep learning. Bio-Keras is TensorFlow, JAX and PyTorch backends compatible and data handling depends on the Numpy.

The comprehensive annotation of human genomic variants is crucial for the biological interpretation and association of these variants with disease phenotypes. Various tools have been developed to annotate variants using established reference databases. In addition to the significant advancements in open-source genomics, parallel initiatives have been undertaken within the industry to facilitate variant identification and annotation. evaiutilities enhances these capabilities by offering a scalable solution for population-level analysis through a command line, a web interface, and a desktop interface.