Szepesi-Nagy, I. | Search bioRxiv

bioRxiv2026

Integrated Interactomics Reveals Novel Protein Associations: The FOXA1-PBX1 Complex as a Case Study

Protein-protein interactions (PPIs) are fundamental to cellular signaling networks, yet many remain undetected due to technical limitations of individual affinity purification approaches. To address this, we systematically mapped the interaction landscapes of six regulatory proteins involved in cell proliferation, immunity, and inflammation, including three transcription factors (TFs) and three kinases. We implemented an integrated proteomics workflow that combined four complementary affinity purification strategies: native immunoprecipitation, two crosslinking-assisted capture methods, and proximity labeling. Combining these approaches revealed distinct yet overlapping interaction profiles, uncovered numerous previously unreported interactors not reliably detected by individual methods, and robustly recovered known interactions while substantially extending PPI networks. Despite method-specific differences at the protein level, functional enrichment analyses showed strong convergence on coherent biological pathways. Biochemical approaches validated most of the previously unreported interactions, including putative weak and transient complexes stabilized by crosslinking. Functional assays revealed a previously unrecognized physical interaction between FOXA1 and PBX1 TFs and demonstrated their cooperative regulation of transcriptional programs and cell fitness in estrogen receptor (ER) positive breast cells. We propose that the FOXA1-PBX1 complex could represent a higher-order regulatory node integrating chromatin accessibility and ER-driven transcriptional output. HIGHLIGHTSO_LIComplementary affinity purification strategies uncover putative weak and transient protein-protein interactions C_LIO_LIFunctional pathway convergence validates biologically coherent interactome expansion C_LIO_LIBiochemical validations confirm unreported interactions C_LIO_LIFunctional validation studies identify a FOXA1-PBX1 pioneer factor complex that regulates estrogen receptor transcriptional programs C_LI Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=191 HEIGHT=200 SRC="FIGDIR/small/738938v1_ufig1.gif" ALT="Figure 1"> View larger version (41K): org.highwire.dtl.DTLVardef@193cc1borg.highwire.dtl.DTLVardef@3d068dorg.highwire.dtl.DTLVardef@791d5corg.highwire.dtl.DTLVardef@1769ce7_HPS_FORMAT_FIGEXP M_FIG C_FIG

biochemistry↗

bioRxiv2023

ProkBERT Family: Genomic Language Models for Microbiome Applications

Machine learning offers transformative capabilities in microbiology and microbiome analysis, deciphering intricate microbial interactions, predicting functionalities, and unveiling novel patterns in vast datasets. This enriches our comprehension of microbial ecosystems and their influence on health and disease. However, the integration of machine learning in these fields contends with issues like the scarcity of labeled datasets, the immense volume and complexity of microbial data, and the subtle interactions within microbial communities. Addressing these challenges, we introduce the ProkBERT model family. Built on transfer learning and self-supervised methodologies, ProkBERT models capitalize on the abundant available data, demonstrating adaptability across diverse scenarios. The models learned representations align with established biological understanding, shedding light on phylogenetic relationships. With the novel Local Context-Aware (LCA) tokenization, the ProkBERT family overcomes the context size limitations of traditional transformer models without sacrificing performance or the information rich local context. In bioinformatics tasks like promoter prediction and phage identification, ProkBERT models excel. For promoter predictions, the best performing model achieved an MCC of 0.74 for E. coli and 0.62 in mixed-species contexts. In phage identification, they all consistently outperformed tools like VirSorter2 and DeepVirFinder, registering an MCC of 0.85. Compact yet powerful, the ProkBERT models are efficient, generalizable, and swift. They cater to both supervised and unsupervised tasks, providing an accessible tool for the community. The models are available on GitHub and HuggingFace.

bioinformatics↗