Biology subjects

Papanikolaou, A.

Publications and source records attributed to Papanikolaou, A..

8 recordsLinked to original sources

bioRxiv2025

A Simple Generative Model for the Prediction of T-Cell Receptor - Peptide Binding in T-Cell Therapy for Cancer

ObjectiveTo develop a deep learning model capable of predicting epitope peptides recognized by specific CDR3 (Complementarity-Determining Region 3) sequences of T-cell receptors (TCRs) in the context of Major Histocompatibility Complex (MHC) molecules, addressing the challenges of incomplete datasets and the need for novel sequence generation in adoptive T-cell therapy for cancer. MethodsWe implemented a sequence to sequence generative model named "GRIP" (Generative Reconstruction of antIgen Peptides) using a Long Short-Term Memory (LSTM) network with attention mechanisms. The model was trained and validated on publicly available datasets, employing data balancing, label smoothing, and dynamic learning rate scheduling to enhance performance and generalization. Accuracy was assessed at the amino acid level. ResultsThe model achieved a training accuracy of 97% and a test accuracy of 85% for predicting epitope sequences at the amino acid level. Probabilistic sequence generation allowed GRIP to produce biologically plausible epitope sequences, even for unseen CDR3 inputs. Attention-based interpretability provided insights into the models focus on critical sequence elements. The model outperformed existing approaches in handling data imbalance and generalization to novel epitopes. ConclusionGRIP offers a novel solution to the TCR-epitope binding problem by generating potential epitope sequences instead of matching to known data, addressing a fundamental gap in existing models. This approach has significant implications for personalized immunotherapy, facilitating the design of targeted T-cell therapies for cancer.

bioinformatics↗

bioRxiv2021

Visual Motion Coherence Responses in Human Visual Cortex

1Random dot kinematograms (RDKs) have recently been used to train subjects with cortical scotomas to perform direction of motion discrimination, partially restoring visual motion perception. To study recovery of visual perception, it is important to understand how visual areas in normal subjects and subjects with cortical scotomas respond to RDK stimuli. Studies in normal subjects have shown that Blood Oxygen Level Dependent (BOLD) responses in human area hV5/MT+ increase monotonically with coherence, in general agreement with electrophysiology studies in primates. However, RDK responses in prior studies were obtained while the subject was performing fixation, not a motion discrimination condition. Furthermore, BOLD responses were gauged against a baseline condition of uniform illumination or static dots, potentially decreasing the specificity of responses for the spatial integration of local motion signals (motion coherence). Here we revisit this question starting from a baseline RDK condition of no coherence, thereby isolating the component of BOLD response due specifically to the spatial integration of local motion signals. In agreement with prior studies, we found that responses in area hV5/MT+ of healthy subjects were monotonically increasing when subjects fixated without performing a motion discrimination task. In contrast, when subjects were performing an RDK direction of motion discrimination task, responses in area hV5/MT+ remained flat, changing minimally, if at all, as a function of motion coherence. A similar pattern of responses was seen in area hV5/MT+ of subjects with dense cortical scotomas performing direction of motion discrimination for RDKs presented inside the scotoma. Passive RDK presentation within the scotoma elicited no significant hV5/MT+ responses. These observations shed further light on how visual cortex responses behave as a function of motion coherence, helping to prepare the ground for future studies using these methods to study visual system recovery after injury.

neuroscience↗

bioRxiv2021

Altered low frequency brain rhythms precede changes in gamma power during tauopathy

Alzheimers disease and other dementias are associated with disruptions of electrophysiological brain activity, including low frequency and gamma rhythms. Many of these dementias are also associated with the malfunction of the membrane associated protein tau. Tauopathy disrupts neuronal function and the stability of synapses and is a key driver of neurodegeneration. Here we ask how brain rhythms are affected by tauopathy, at different stages of its progression. We performed local field potential recordings from visual cortex of rTg4510 and control animals at early stages of neurodegeneration (5 months) and at a more advanced stage where pathology is evident (8 months). We measured brain activity in the presence or absence of external visual stimulation, and while monitoring pupil diameter and locomotion to establish animal behavioural states. At 5 months, before substantial pathology, we found an increase in low frequency rhythms during resting state in tauopathic animals. This was because tauopathic animals entered intermittent periods of increased neural synchronisation, where activity across a wide band of low frequencies was strongly correlated. At 8 months, when the degeneration was more advanced, the increased synchronisation and low frequency power was accompanied by a reduction in power in the gamma range, with diverse effects across different components of the gamma rhythm. Our results indicate that slower rhythms are impaired earlier than gamma rhythms in tauopathy, suggesting that electrophysiological measurements can indicate both the presence and progression of tauopathic degeneration.

neuroscience↗

bioRxiv2020

Plasticity in visual cortex is disrupted in a mouse model of tauopathy and neurodegeneration

Neurodegeneration is a hallmark of many dementias and is thought to underlie a progressive impairment of neural plasticity. How neurodegeneration affects plasticity in neural circuits is not known. We therefore characterised the impact of tau-driven neurodegeneration on plasticity in the visual system, where normal function is well understood. We studied a very simple form of visual plasticity that allowed us to track both long timescales (across days) and shorter timescales (over minutes). We recorded the local field potential in the primary visual cortex of rTg4510 transgenic mice, a mouse model of tauopathy, while animals were repeatedly exposed to the same stimulus over the course of 9 days. We studied animals at early stages of neurodegeneration (5 months old) and at a more advanced stage where pathology is evident (8 months). We found that both short- and long-term visual plasticity were already disrupted at early stages of neurodegeneration, and were further reduced in older animals, such that it was abolished in mice expressing the mutant tau. Additionally, we found that visually evoked behaviours were disrupted in both younger and older mice expressing the mutant tau. Our results show that visual cortical plasticity and visually evoked behaviours are disrupted in the rTg4510 model of tauopathy, even at early stages of neurodegeneration. This simple measure of neural plasticity may help understand how neurodegeneration disrupts neural circuits, and offers a translatable platform for detection and tracking of the disease. HighlightsO_LIVisual plasticity is disrupted at early stages of neurodegeneration in rTg4510 mice C_LIO_LIVisual plasticity is reduced in older animals, particularly during neurodegeneration C_LIO_LIInstinctive visual behaviours are reduced in neurodegeneration C_LIO_LIShort-term visual plasticity is reduced in neurodegeneration C_LI

neuroscience↗

bioRxiv2020

BonVision - an open-source software to create and control visual environments

Real-time rendering of closed-loop visual environments is necessary for next-generation understanding of brain function and behaviour, but is prohibitively difficult for non-experts to implement and is limited to few laboratories worldwide. We developed BonVision as an easy-to-use open-source software for the display of virtual or augmented reality, as well as standard visual stimuli. As the architecture is based on the open-source Bonsai graphical programming language, BonVision benefits from native integration with experimental hardware. BonVision therefore enables easy implementation of closed-loop experiments, including real-time interaction with deep neural networks and communication with behavioural and physiological measurement and manipulation devices.

neuroscience↗

bioRxiv2018

Detection of acute TLR-7 agonist-induced hemorrhagic myocarditis in mice by refined multi-parametric quantitative cardiac MRI.

BACKGROUNDHemorrhagic myocarditis is a potentially fatal complication of excessive levels of systemic inflammation. It has been reported in viral infection, but is also possible in systemic autoimmunity. Cardiac magnetic resonance (CMR) imaging is the current gold standard for non-invasive detection of suspected inflammatory damage to the heart and changes in T1 and T2 relaxation times are commonly used to detect edema associated with immune cell infiltration and fibrosis. These measurements also form the basis of the Lake Louise Criteria, which define a framework for the CMR-based diagnosis of myocarditis. However, they do not take into account the possibility of hemorrhage leading to tissue iron deposition which strongly influences T1 and T2 measurements and may complicate interpretation based on these two parameters only.\n\nMETHODSSystemic inflammation was induced in CFN mice by application of the TLR-7 agonist Resiquimod. Histopathology was performed on heart sections to assess immune cell infiltration (Hematoxylin & Eosin), fibrosis (PicoSirius Red) and tissue iron deposition (Perls Prussian Blue). A multi-parametric cardiac MRI tissue mapping approach measuring T1, T2, and T2* relaxation times was established to non-invasively identify these parameters in this small rodent model.\n\nRESULTSResiquimod-treated mice developed severe thrombocytopenia and hemorrhagic myocarditis. We identified patches of cardiac hemorrhage based on the presence of two major MRI phenotypes. Increased T2 with normal T1 and T2* values correlated with infiltration/edema only, while decreased T1, T2, and T2* values identify areas with infiltration/edema in the presence of iron indicating hemorrhagic myocarditis.\n\nCONCLUSIONWe show that over-activation of the TLR-7 pathway by Resiquimod treatment of CFN mice induces an early immune response reminiscent of excessive systemic inflammation due to viral infection. This causes internal bleeding which manifests most prominently as severe hemorrhagic myocarditis. We optimized a comprehensive noninvasive in vivo MRI approach based on quantitative measurement of T1, T2 and T2* to demonstrate the presence of diffuse myocardial edema, infiltration and iron deposition without the need of contrast agent administration. We propose that adding quantitative T2* mapping to CMR protocols for detection of myocarditis will improve diagnostic sensitivity and interpretation of disease mechanisms.

physiology↗

bioRxiv2018

Measuring single neuron visual receptive field sizes by fMRI

The non-invasive measurements of neuronal receptive field (RF) properties in-vivo allow a detailed understanding of brain organization as well as its plasticity by longitudinal following of potential changes. Visual RFs measured invasively by electrophysiology in animal models have traditionally provided a great extent of our current knowledge about the visual brain and its disorders. Voxel based estimates of population RF (pRF) by functional magnetic resonance imaging (fMRI) in humans revolutionized the field and have been used extensively in numerous studies. However, current methods cannot estimate single-neuron RF sizes as they reflect large populations of neurons with individual RF scatter. Here, we introduce a new approach to estimate RF size using spatial frequency selectivity to checkerboard patterns. This method allowed us to obtain non-invasive, single-unit, RF estimates in human V1 for the first time. These estimates were significantly smaller compared to prior pRF methods. Further, fMRI and electrophysiological experiments in non-human primates demonstrated an exceptional match validating the approach.

neuroscience↗

bioRxiv2018

Organization of Area hV5/MT+ in Subjects with Homonymous Visual Field Defects

Damage to the primary visual cortex (V1) leads to a visual field loss (scotoma) in the retinotopically corresponding part of the visual field. Nonetheless, a small amount of residual visual sensitivity persists within the blind field. This residual capacity has been linked to activity observed in the middle temporal area complex (V5/MT+). However, it remains unknown whether the organization of hV5/MT+ changes following V1 lesions. We studied the organization of area hV5/MT+ of five patients with dense homonymous defects in a quadrant of the visual field as a result of partial V1+ or optic radiation lesions. To do so, we developed a new method, which models the boundaries of population receptive fields directly from the BOLD signal of each voxel in the visual cortex. We found responses in hV5/MT+ arising inside the scotoma for all patients and identified two possible sources of activation: 1) responses might originate from partially lesioned parts of area V1 corresponding to the scotoma, and 2) responses can also originate independent of area V1 input suggesting the existence of functional V1-bypassing pathways. Apparently, visually driven activity observed in hV5/MT+ is not sufficient to mediate conscious vision. More surprisingly, visually driven activity in corresponding regions of V1 and early extrastriate areas including hV5/MT+ did not guarantee visual perception in the group of patients with post-geniculate lesions that we examined. This suggests that the fine coordination of visual activity patterns across visual areas may be an important determinant of whether visual perception persists following visual cortical lesions.

neuroscience↗