Biology subjects

Pham, B.

Publications and source records attributed to Pham, B..

6 recordsLinked to original sources

bioRxiv2026

ARCHER: Amortized cross-specimen pose estimation for cryo-electron microscopy

Single-particle cryo-electron microscopy (cryo-EM) pose estimation is traditionally solved anew for each dataset, where iterative refinement is done from scratch while the estimator learns to store the molecule in its weights. In this work, we show that pose inference is a generalizable, specimen-agnostic operation when conditioned explicitly on a reference volume. We introduce ARCHER, an amortized contrastive classifier that models the pose posterior over a discrete rotation grid. Trained across a variety of protein structures, it operates zero-shot without retraining per structure. This transferability is grounded in Fourier-space information mechanics, where all specimen dependence is captured by the reference structure's power spectrum and spatial extent. ARCHER achieves a median angular error of 5.0{degrees} on 100 held-out test structures and 2.5{degrees} on experimental particles, matching dedicated estimators within 0.16[A] in 3D reconstruction. Crucially, downstream conformational signal is preserved. The leading conformational coordinate correlates at 0.97 with deposited benchmarks, faithfully reconstructing free-energy basins and mobile domains. These results overall demonstrate that cryo-EM pose estimation can be generalized across different structures.

biophysics↗

bioRxiv2026

Machine Learning Ensemble Reveals Distinct Molecular Pathways of Retinal Damage in Spaceflown Mice

BackgroundSpaceflight-associated neuro-ocular syndrome (SANS) poses significant risks to astronaut visual health during long-duration missions, yet its underlying molecular mechanisms remain incompletely understood. Oxidative stress and apoptosis are candidate molecular drivers, but their transcriptomic signatures in spaceflight-exposed retinal tissue have not been systematically characterized. MethodsWe applied a machine learning ensemble of linear regression models to predict two ocular phenotypes: 4-hydroxynonenal (4-HNE) immunostaining as a marker of lipid peroxidation-mediated oxidative damage; and TUNEL positivity as a marker of apoptotic cell death. In this observational study, we use bulk retinal gene expression data obtained from a controlled experiment with ground control and spaceflown mice to predict these phenotypes. Gene Ontology pathway enrichment was performed on the most predictive gene sets for each phenotype. ResultsThe 4-HNE phenotype was predicted by genes that converge on membrane-associated pathways, photoreceptor protein modification, synaptic dysfunction, and extracellular matrix dysregulation, including B2m, Tf, Cnga1, mt-Nd1, Snap25, and Efemp1. The genes predicting the TUNEL phenotype revealed a distinct signature emphasizing stress-induced apoptosis, rod photoreceptor degeneration, and endoplasmic reticulum dysfunction, with Ddit4, Nrl, Rom1, Reep6, and Gabarapl1 emerging as central regulators. ConclusionsOxidative lipid peroxidation and apoptotic cell death represent complementary and molecularly distinct pathological mechanisms in spaceflight-exposed murine retinal tissue. The gene signatures provide a putative molecular framework for developing noninvasive biomarkers and therapeutic targets to monitor and protect astronaut visual health during long-duration and deep-space missions.

bioinformatics↗

bioRxiv2025

Comparison of oral and gut microbiome highlights role of oral bacteria in systemic inflammation in HIV

BackgroundChronic HIV-1 infection is associated with increased inflammation-related comorbidities, despite effective viral suppression with antiretroviral therapy. While the role of the gut microbiome in inflammation is well-studied, the contribution of the oral microbiome remains less clear. This study investigates the relationship between the oral and gut microbiomes in driving systemic inflammation in persons with HIV. MethodsThis cross-sectional study utilized archived samples from 198 participants (99 with HIV and 99 without HIV). Oral microbiome composition was analyzed via 16S rRNA sequencing and systemic inflammatory biomarkers were measured using multiplex assays. Gut microbiome data from previous studies were integrated for comparative analyses. Bacterial inflammatory potential was assessed through in vitro co-culture and epithelial barrier permeability assays. ResultsThe oral microbiome in HIV was characterized by increased Veillonella, Capnocytophaga, and Megasphaera, and several decreased genera including Fusobacterium. Using PERMANOVA, we found that the oral microbiome was a significant driver of cytokine variation in HIV compared to the gut microbiome, and identified specific associations with oral Veillonella and Megasphaera. We found no differences in anti-Veillonella parvula serum IgG by HIV status, but IgG titers did correlate with microbial translocation markers sCD14 and LBP in HIV. In vitro studies demonstrated that Veillonella parvula increased oral epithelial barrier permeability and induced monocyte activation. ConclusionsThe oral microbiome, particularly Veillonella parvula, may contributes to systemic inflammation in HIV through mechanisms involving epithelial barrier disruption, oral translocation, and monocyte activation.

microbiology↗

bioRxiv2023

Chitinase 3-like-1 (CHI3L1) in the Pathogenesis of Epidermal Growth Factor Receptor Mutant Non-Small Cell Lung Cancer

Non-small cell lung cancer (NSCLC) accounts for 85% of all lung cancers. In NSCLC, 10-20% of Caucasian patients and 30-50% of Asian patients have tumors with activating mutations in the Epidermal Growth Factor Receptor (EGFR). A high percentage of these patients exhibit favorable responses to treatment with tyrosine kinase inhibitors (TKI). Unfortunately, a majority of these patients develop therapeutic resistance with progression free survival lasting 9-18 months. The mechanisms that underlie the tumorigenic effects of EGFR and the ability of NSCLC to develop resistance to TKI therapies, however, are poorly understood. Here we demonstrate that CHI3L1 is produced by EGFR activation of normal epithelial cells, transformed epithelial cells with wild type EGFR and cells with cancer-associated, activating EGFR mutations. We also demonstrate that CHI3L1 auto-induces itself and feeds back to stimulate EGFR and its ligands. Highly specific antibodies against CHI3L1 (anti-CHI3L1/FRG) and TKI, individually and in combination, abrogated the effects of EGFR activation on CHI3L1 and the ability of CHI3L1 to stimulate the EGFR axis. Anti-CHI3L1 also interacted with osimertinib to reverse TKI therapeutic resistance and induce tumor cell death and inhibit pulmonary metastasis while stimulating tumor suppressor genes including KEAP1. CHI3L1 is a downstream target of EGFR that feeds back to stimulate and activate the EGFR axis. Anti-CHI3L1 is an exciting potential therapeutic for EGFR mutant NSCLC, alone and in combination with osimertinib or other TKIs.

cancer biology↗

bioRxiv2021

Expanding the Molecular Alphabet of DNA-Based Data Storage Systems with Neural Network Nanopore Readout Processing

DNA is a promising next-generation data storage medium, but the recording latency and synthesis cost of oligos using the four natural nucleotides remain high. Here, we describe an improved DNA-based storage system that uses an extended 11-letter molecular alphabet combining natural and chemically modified nucleotides. Our extended-alphabet molecular storage paradigm offers a nearly two-fold increase in storage density and potentially the same order of reduction in the recording time. Experimental results involving a library of 77 custom-designed hybrid sequences reveal that one can readily detect and discriminate different combinations and orders of monomers via MspA nanopores. Furthermore, a neural network architecture designed to classify raw current signals generated by Oxford Nanopore Technologies sequencing ensures an average accuracy exceeding 60%, which is 39 times higher than that of random guessing. Molecular dynamics simulations reveal that the majority of modified nucleotides do not induce dramatic disruption of the DNA double helix, making the extended alphabet system potentially compatible with PCR-based random access data retrieval. The methodologies proposed provide a forward path for new implementations of molecular recorders.

synthetic biology↗

bioRxiv2018

Characterizing the 3D structure and dynamics of chromosomes and proteins in a common contact matrix framework

Any conformational ensemble of biopolymers, whether they are proteins or chromosomes, can be described using contact matrices derived from experimental or computational studies. One powerful approach to extract meaningful information from these contact matrices is to perform principal component analysis (PCA) on the covariance matrix of the contact data. Indeed, PCA on Hi-C chromosome contact matrices has revealed the spatial segregation of active and inactive chromatin. Separately, PCA on contact matrices from snapshots of protein conformations has characterized correlated fluctuations of protein domains. However, despite the similarities of these data and analyses, there has been little synergy between the PCA approaches and the comparison of resulting biological insights obtained for protein and chromosome structures. We note that, to date, different styles of analyses were applied exclusively to each biomolecule type: explicit contact correlation analysis (E-PCA) for proteins and implicit contact correlation (I-PCA) for chromosomes. In this work, we compare the results of applying both methods to both classes of biopolymers. While I-PCA reveals only average features of 3D chromosome structure, we find that applying E-PCA to an ensemble of chromosome structures from microscopy data reveals the dominant motion (concerted fluctuation) of the chromosome. Applying E-PCA to Hi-C data across the human blood cell lineage isolates the aspects of chromosome structure that most strongly differentiate cell types. Conversely, when we apply I-PCA to simulation snapshots of two protein complexes, the major component reports the consensus features of the structure, while the previously applied E-PCA characterizes correlated deviations from the mean structure.

biophysics↗