In bioinformatics, machine learning methods have been used to predict features embedded in the sequences. In contrast to what is generally assumed, machine learning approaches can also provide new insights into the underlying biology. Here, we demonstrate this by presenting TargetP 2.0, a novel state-of-the-art method to identify N-terminal sorting signals, which direct proteins to the secretory pathway, mitochondria, and chloroplasts or other plastids. By examining the strongest signals from the attention layer in the network, we find that the second residue in the protein, that is, the one following the initial methionine, has a strong influence on the classification. We observe that two-thirds of chloroplast and thylakoid transit peptides have an alanine in position 2, compared with 20% in other plant proteins. We also note that in fungi and single-celled eukaryotes, less than 30% of the targeting peptides have an amino acid that allows the removal of the N-terminal methionine compared with 60% for the proteins without targeting peptide. The importance of this feature for predictions has not been highlighted before.

Single-cell RNA sequencing (scRNA-seq) technology has significantly advanced our understanding of the diversity of cells and how this diversity is implicated in diseases. Yet, translating these findings across various scRNA-seq datasets poses challenges due to technical variability and dataset-specific biases. To overcome this, we present a novel approach that employs both an LLM-based framework and explainable machine learning to facilitate generalization across single-cell datasets and identify gene signatures to capture disease-driven transcriptional changes. Our approach uses scBERT, which harnesses shared transcriptomic features among cell types to establish consistent cell-type annotations across multiple scRNA-seq datasets. Additionally, we employed a symbolic regression algorithm to pinpoint highly relevant, yet minimally redundant models and features for inferring a cell type’s disease state based on its transcriptomic profile. We ascertained the versatility of these cell-specific gene signatures across datasets, showcasing their resilience as molecular markers to pinpoint and characterize disease-associated cell types. The validation was carried out using four publicly available scRNA-seq datasets from both healthy individuals and those suffering from ulcerative colitis (UC). This demonstrates our approach’s efficacy in bridging disparities specific to different datasets, fostering comparative analyses. Notably, the simplicity and symbolic nature of the retrieved gene signatures facilitate their interpretability, allowing us to elucidate underlying molecular disease mechanisms using these models.

Transposable elements are an abundant source of transcription factor binding sites, and favorable genomic integration may lead to their recruitment by the host genome for gene regulatory functions. However, it is unclear how frequent co-option of transposable elements as regulatory elements is, to which regulatory programs they contribute and how they compare to regulatory elements devoid of transposable elements. Here, we report a transcription initiation-centric, in-depth characterization of the transposon-derived regulatory landscape of mouse embryonic stem cells. We demonstrate that a substantial number of transposable element insertions, in particular endogenous retroviral elements, are associated with open chromatin regions that are divergently transcribed into unstable RNAs in a cell-type specific manner, and that these elements contribute to a sizable proportion of active enhancers and gene promoters. We further show that transposon subfamilies contribute differently and distinctly to the pluripotency regulatory program through their repertoires of transcription factor binding site sequences, shedding light on the formation of regulatory programs and the origins of regulatory elements.

Genetic and environmental exposures cause variability in gene expression. Although most genes are affected in a population, their effect sizes vary greatly, indicating the existence of regulatory mechanisms that could amplify or attenuate expression variability. Here, we investigate the relationship between the sequence and transcription start site architectures of promoters and their expression variability across human individuals. We find that expression variability can be largely explained by a promoter’s DNA sequence and its binding sites for specific transcription factors. We show that promoter expression variability reflects the biological process of a gene, demonstrating a selective trade-off between stability for metabolic genes and plasticity for responsive genes and those involved in signaling. Promoters with a rigid transcription start site architecture are more prone to have variable expression and to be associated with genetic variants with large effect sizes, while a flexible usage of transcription start sites within a promoter attenuates expression variability and limits genotypic effects. Our work provides insights into the variable nature of responsive genes and reveals a novel mechanism for supplying transcriptional and mutational robustness to essential genes through multiple transcription start site regions within a promoter.

Dysfunction of regulatory elements through genetic variants is a central mechanism in the pathogenesis of disease. To better understand disease etiology, there is consequently a need to understand how DNA encodes regulatory activity. Deep learning methods show great promise for modeling of biomolecular data from DNA sequence but are limited to large input data for training. Here, we develop ChromTransfer, a transfer learning method that uses a pre-trained, cell-type agnostic model of open chromatin regions as a basis for fine-tuning on regulatory sequences. We demonstrate superior performances with ChromTransfer for learning cell-type specific chromatin accessibility from sequence compared to models not informed by a pre-trained model. Importantly, ChromTransfer enables fine-tuning on small input data with minimal decrease in accuracy. We show that ChromTransfer uses sequence features matching binding site sequences of key transcription factors for prediction. Together, these results demonstrate ChromTransfer as a promising tool for learning the regulatory code.

The identification of predictive biomarker signatures from omics data for clinical applications is an active area of research. Recent developments in assay technologies and machine learning (ML) methods have led to significant improvements in predictive performance. However, most high-performing ML methods suffer from complex architectures and lack interpretability. Here, we present the application of a novel symbolic-regression-based algorithm, the QLattice, on a selection of clinical omics data sets. This approach generates parsimonious high-performing models that can both predict disease outcomes and reveal putative disease mechanisms. Due to their high performance, simplicity and explicit functional form, these biomarker signatures can be readily explained, thereby making them attractive tools for high-stakes applications in primary care, clinical decision making and patient stratification.