Search PubMed⌕ Search

Biomedical subjects

June L Round

Publications and source records attributed to June L Round.

3 recordsLinked to original sources

LAMBDA: a prophage detection benchmark for genomic language models.

Transformer-based genomic sequence models represent an emerging frontier in computational biology. Yet, their embeddings have not yet shown the same level of predictive power as natural and protein language models, highlighting a gap between current implementations and theoretical promise. Existing benchmarks for DNA language models primarily focus on classifying regulatory elements in eukaryotic genomes, leaving open the fundamental question of whether these models learn sequence-level features across whole genomes. We introduce LAMBDA, a benchmark designed to rigorously evaluate genome language model embeddings through phage-bacteria sequence discrimination across four categories of increasing complexity: probing tasks, fine-tuning assessments, diagnostic tests, and genome-wide prophage detection. Our comprehensive analysis of current genomic language models provides insight into the importance of training data selection relative to model size, the need for domain-specific training, and the capabilities and limitations of genomic language models for detecting prophage sequences. This benchmark represents a challenging genomic annotation task in the bacterial domain and addresses a key computational problem with direct relevance to microbiology and medicine.

Prophages↗

Scaffold protein Dlgh1 coordinates alternative p38 kinase activation, directing T cell receptor signals toward NFAT but not NF-kappaB transcription factors.

Tyrosine kinases couple the T cell receptor (TCR) to discrete signaling cascades, each of which is capable of inducing a distinct functional outcome. Precisely how TCR signals are channeled toward specific targets remains unclear. TCR stimulation triggers 'alternative' activation of the mitogen-activated protein kinase p38, whereby the Lck and Zap70 tyrosine kinases directly activate p38. Here we report that alternatively activated p38 associated with the Dlgh1 MAGUK scaffold protein. 'Knockdown' of Dlgh1 expression blocked TCR-induced activation of p38 and the transcription factor NFAT but not of the mitogen-activated protein kinase Jnk or transcription factor NF-kappaB. A Dlgh1 mutant incapable of binding p38 failed to activate NFAT. Along with reports that the CARMA1 MAGUK scaffold protein coordinates activation of Jnk and NF-kappaB but not of p38 or NFAT, our findings identify MAGUK scaffold proteins as 'orchestrators' of TCR signal specificity.

Adaptor Proteins, Signal Transducing↗

Dlgh1 coordinates actin polymerization, synaptic T cell receptor and lipid raft aggregation, and effector function in T cells.

Lipid raft membrane compartmentalization and membrane-associated guanylate kinase (MAGUK) family molecular scaffolds function in establishing cell polarity and organizing signal transducers within epithelial cell junctions and neuronal synapses. Here, we elucidate a role for the MAGUK protein, Dlgh1, in polarized T cell synapse assembly and T cell function. We find that Dlgh1 translocates to the immune synapse and lipid rafts in response to T cell receptor (TCR)/CD28 engagement and that LckSH3-mediated interactions with Dlgh1 control its membrane targeting. TCR/CD28 engagement induces the formation of endogenous Lck-Dlgh1-Zap70-Wiskott-Aldrich syndrome protein (WASp) complexes in which Dlgh1 acts to facilitate interactions of Lck with Zap70 and WASp. Using small interfering RNA and overexpression approaches, we show that Dlgh1 promotes antigen-induced actin polymerization, synaptic raft and TCR clustering, nuclear factor of activated T cell activity, and cytokine production. We propose that Dlgh1 coordinates TCR/CD28-induced actin-driven T cell synapse assembly, signal transduction, and effector function. These findings highlight common molecular strategies used to regulate cell polarity, synapse assembly, and transducer organization in diverse cellular systems.

Actins↗