Biology subjects

Matuleviciute, R.

Publications and source records attributed to Matuleviciute, R..

4 recordsLinked to original sources

bioRxiv2026

Exploring noise in TCR specificity data through supervised learning

Accurate prediction of T cell receptor (TCR) specificity can potentially accelerate development of novel immunotherapies, yet public databases used to train these models contain substantial label noise. This uncertainty complicates model evaluation and limits the reliability of predicted TCR-pMHC interactions. In this work, we present a simple supervised machine learning denoising scheme to identify low-quality labels in TCR specificity data. We evaluated the framework using the TCRvdb validation dataset and further corroborated it via structural modeling with AlphaFold 3. Here, our denoising approach successfully identified non-validated TCRs, achieving an average MCC of 0.47 for filtering non-validated examples while retaining high sensitivity for validated examples. Applying our denoising approach to the complete NetTCR training dataset, we observed that data associated with known noisy attributes were significantly enriched in the filtered subset. Furthermore, TCR-pMHC complexes identified as noise consistently produced lower AlphaFold 3 confidence scores, providing independent evidence of misannotation. Experimental validation of 30 selected TCRs further supported the approach, with 80% of TCRs predicted as noise failing to demonstrate functional reactivity. Our findings demonstrate that current performance metrics may be obscured by dataset inaccuracies and provide a method for refining training data in the absence of gold-standard benchmarks.

bioinformatics↗

bioRxiv2023

Transcriptional and Cellular Response of hiPSC-derived Microglia-Neural Progenitor Co-Cultures Exposed to IL-6

Elevated interleukin (IL-)6 levels during prenatal development have been linked to increased risk for neurodevelopmental disorders (NDD) in the offspring, but the mechanism remains unclear. Human-induced pluripotent stem cell (hiPSC) models offer a valuable tool to study the effects of IL-6 on features relevant for human neurodevelopment in vitro. We previously reported that hiPSC-derived microglia-like cells (MGLs) respond to IL-6, but neural progenitor cells (NPCs) in monoculture do not. Therefore, we investigated whether co-culturing hiPSC-derived MGLs with NPCs would trigger a cellular response to IL-6 stimulation via secreted factors from the MGLs. Using N=4 donor lines without psychiatric diagnosis, we first confirmed that NPCs can respond to IL-6 through trans-signalling when recombinant IL-6Ra is present, and that this response is dose-dependent. MGLs secreted soluble IL-6R, but at lower levels than found in vivo and below that needed to activate trans-signalling in NPCs. Whilst transcriptomic and secretome analysis confirmed that MGLs undergo substantial transcriptomic changes after IL-6 exposure and subsequently secrete a cytokine milieu, NPCs in co-culture with MGLs exhibited a minimal transcriptional response. Furthermore, there were no significant cell fate-acquisition changes when differentiated into post-mitotic cultures, nor alterations in synaptic densities in mature neurons. These findings highlight the need to investigate if trans-IL-6 signalling to NPCs is a relevant disease mechanism linking prenatal IL-6 exposure to increased risk for psychiatric disorders. Moreover, our findings underscore the importance of establishing more complex in vitro human models with diverse cell types, which may show cell-specific responses to microglia-released cytokines to fully understand how IL-6 exposure may influence human neurodevelopment.

neuroscience↗

bioRxiv2022

Acute IL-6 exposure triggers canonical IL-6R signalling in hiPSC microglia, but not neural progenitor cells.

BackgroundExposure to elevated interleukin (IL)-6 levels in utero is consistently associated with increased risk for psychiatric disorders with a putative neurodevelopmental origin, such as schizophrenia (SZ) and autism spectrum condition (ASC). Although rodent models provide causal evidence for this association, we lack a detailed understanding of the cellular and molecular mechanisms in human model systems. To close this gap, we characterised the response of hiPSC-derived microglia-like cells (MGL) and neural progenitor cells (NPCs) to IL-6 in monoculture. ResultsWe observed that human forebrain NPCs did not respond to acute IL-6 exposure in monoculture at both a protein and transcript level due to the absence of IL-6Ra expression and sIL-6Ra secretion. By contrast, acute IL-6 exposure resulted in STAT3 phosphorylation and increased IL-6, JMJD3 and IL-10 expression in MGL, confirming activation of canonical IL-6R signalling. Bulk RNAseq identified 156 upregulated genes (FDR <0.05) in MGL following acute IL-6 exposure, including IRF8, REL, HSPA1A/B and OXTR, which significantly overlapped with an upregulated gene set from post-mortem brain tissue from individuals with schizophrenia. Acute IL-6 stimulation significantly increased MGL motility suggestive of a gain of surveillance function, consistent with gene ontology pathways highlighted from the RNAseq data. Finally, MGLs displayed elevated CCL1, CXCL1, MIP-1A/B, IL-8, IL-13, IL-16, IL-18, MIF and Serpin-E1 secretion post 3h and 24h IL-6 exposure. ConclusionOur data provide evidence for cell specific effects of acute IL-6 exposure in a human model system and strongly suggest microglia-NPC co-culture models are required to study how IL-6 influences human cortical neural progenitor cell development in vitro.

neuroscience↗

bioRxiv2021

Interferon-γ Exposure of Human iPSC-derived Neurons Alters Major Histocompatibility Complex I and Synapsin I Protein Expression

Human epidemiological data links maternal immune activation during gestation with increased risk for neurodevelopmental disorders including schizophrenia. Animal models of maternal immune activation (MIA) provide causal evidence for this association and strongly suggest that inflammatory cytokines act is a critical link between maternal infection and aberrant offspring brain and behavior development. This includes evidence for reduced synapse formation, consistent with post-mortem and in vivo evidence of reduced synaptic density in schizophrenia. However, to what extent specific cytokines are necessary and sufficient for these effects remains unclear. Using a human cellular model, we recently demonstrated that acute exposure to interferon-{square} (IFN{square}) recapitulates molecular and cellular phenotypes associated with neurodevelopmental disorders. Here, we extend this work to test whether IFN{square} affects synapse formation in an induced neuron model that generates forebrain glutamatergic neurons. Using immunocytochemistry and quantitative PCR, we demonstrate that acute IFN{square} exposure results in significantly increased MHCI expression at the mRNA and protein level. Furthermore, acute IFN{square} exposure decreases synapsin I protein in neurons but does not affect synaptic gene mRNA levels. Interestingly, complement component 4A (C4A) mRNA is also significantly increased following acute IFN{square} exposure. This study builds on our previous work by showing that IFN{square}-mediated disruption of relevant synaptic proteins can occur at early stages of synapse formation, potentially contributing to neurodevelopmental disorder phenotypes such as schizophrenia.

neuroscience↗