Biology subjects

Lai, B.

Publications and source records attributed to Lai, B..

15 recordsLinked to original sources

bioRxiv2026

DecoderTCR: Compositional Pretraining and Entropy-Guided Decoding for TCR-pMHC Interactions

Modeling recognition between T-cell receptors (TCRs) and peptide-MHC (pMHC) complexes is a fundamental challenge in computational immunology, constrained by sparse paired interaction data relative to abundant unpaired sequences. We introduce DecoderTCR, a masked language model framework that addresses this through two contributions: (1) a compositional continual pre-training curriculum that learns component representations from marginal data before refining cross-chain dependencies from limited pairs, and (2) Iterative Entropy-Guided Refinement (IEGR), a non-autoregressive decoding algorithm that resolves high-confidence positions first to provide context for uncertain regions. On held-out benchmarks, DecoderTCR achieves 0.96 AUROC for zero-shot pMHC binding prediction and 0.76 AUROC for epitope-specific TCR recognition, approaching supervised baselines without epitope-specific training. Learned representations recover structural contacts without coordinate supervision, and generated sequences exhibit realistic recombination statistics. Experimental validation reveals a prediction-generation gap: strong discrimination does not yet yield reliable generation, highlighting an open challenge for the field.

bioinformatics↗

bioRxiv2025

Mesoderm-specific transcript reduces ciliary sphingomyelin levels to promote tendon stem/progenitor cells osteochondrogenesis in traumatic heterotopic ossification

Traumatic heterotopic ossification (tHO) is a musculoskeletal disorder characterized by ectopic bone formation in soft tissues following trauma. However, the etiology and mechanisms of tHO remain unknown, and no treatments are highly effective for tHO. Here, by performing single-cell RNA sequencing (scRNA-seq) on two tHO mouse models, we found that ciliary Hedgehog (Hh) and PI3K-Akt signaling in tendon stem/progenitor cells (TSPCs) is upregulated during tHO development, leading to the activation of GLI family zinc finger 2 (GLI2) transcription factor, which promotes mesoderm-specific transcript (MEST) gene expression. Untargeted lipidomics using liquid chromatography-tandem mass spectrometry (LC-MS/MS) on detached cilia indicated that MEST could reduce ciliary sphingomyelin (SM) levels by inhibiting SM synthesis in TSPCs, creating a positive feedback loop that amplified ciliary Hh signaling, enhancing TSPCs osteochondrogenesis and driving tHO formation. Targeting ciliary genes intraflagellar transport 88 (IFT88) or ADP ribosylation factor like GTPase 3 (ARL3) restored normal TSPCs osteochondrogenesis in vitro and attenuated ectopic bone formation in vivo by suppressing Hh and PI3K-Akt signaling. AAV-mediated MEST inhibition or exogenous SM administration in vivo also alleviates tHO progression. These findings may provide novel insights into tHO pathogenesis and potential therapeutic strategies.

molecular biology↗

bioRxiv2025

Multifunctional bending magnet beamline with a capillary optic for X-ray fluorescence studies of metals in tissue sections

Scanning fluorescence X-ray microscopy lets one non-destructively and quantitatively map the distribution of most biologically-important metals in cells and tissues. For studies on large-scale tissues and organs, a spatial resolution of several micrometers is often sufficient; in this case, bending magnets at synchrotron light sources provide abundant X-ray flux. We describe here the use of bending magnet beamline 8-BM-B at the Advanced Photon Source (APS) with two distinct microscopy stations: a pre-existing one with Kirkpatrick-Baez (KB) mirror optics for slightly higher throughput and the ability to accommodate samples tens of centimeters across, and a new prototype station with an axially-symmetric, single-bounce, capillary optic with slightly less flux, but slightly higher fluence (which affects achievable resolution at low metal concentration) and higher spatial resolution. The KB station provides{delta} res = 10.5 {micro}m spatial resolution at a per-pixel exposure time of tdwell = 100 ms and a fluence per time of 5.8x 107 photons /({micro}m2 {middle dot}s), while the prototype capillary station provides{delta} res = 6.3 {micro}m at tdwell = 50 ms and a fluence per time of 6.1x 107 photons ({micro}m2 {middle dot}s). We used image power spectral density to estimate the achieved spatial resolution{delta} res from individually acquired images, with{delta} res depending-on the optic, the fluorescence signal strength of the sample being imaged, and the method used to process raw fluorescence spectral data.

biophysics↗

bioRxiv2025

Contribution of DNA breathing to physical interactions with transcription factors

Interaction between transcription factors (TFs) and DNA plays a key role in regulating gene expression. It is generally believed that these interactions are controlled through recognition of DNA core motifs by TFs. Nevertheless, several studies pointed out the limitation of this view, in particular, DNA sequence variants influencing TF binding are often located outside of core motifs. One possible explanation is that the physical properties of DNA may play a role in TF-DNA interactions. Recent studies have supported the importance of DNA shape features, especially in flanking regions of core motifs. Another important physical property of DNA is DNA breathing, the spontaneous opening of double-stranded DNA through thermal motions. But there have been few genomic studies of the role of DNA breathing in TF-DNA interactions. In this work, we analyzed in vitro TF-DNA binding data of three TFs and found that DNA breathing features inside or near core motifs are correlated with binding affinity. This suggests that these TFs may prefer locally and temporally melted DNA formed through breathing. We extended the analysis to 44 TFs with in vivo ChIP-seq binding data. We found that for a large proportion of TFs, their breathing features in or near core motifs are associated with binding, but the sign and magnitude of these associations vary substantially across TF families. Altogether, our study supports the hypothesis that DNA breathing features near binding motifs contribute to TF-DNA interactions. Author SummaryProper regulation of when and where genes are expressed is crucial to biological development and function. This process is largely controlled by interaction of transcription factors (TFs) with DNA sequences. The recognition of specific DNA sequences by TFs is important to ensure that only the correct genes are activated. Extensive work has shown that TFs prefer to bind certain DNA sequence patterns of 6-20 bp, known as motifs. However, the structure of DNA molecules may also play a role. In this work, we explored the role of DNA breathing, which refers to spontaneous opening of double strand DNA due to thermal motions. This process creates transient, single-strand "bubbles" in DNA. Through examining TF-DNA binding data of >60 TFs, we found that the propensity of DNA forming bubbles near motifs is often associated with binding affinity of DNA sequence. Interestingly the patterns of these associations seem to vary with TFs. Altogether, our results highlighted the potential of DNA breathing in influencing TF-DNA interactions.

genetics↗

bioRxiv2024

RIG-I and MDA5 cooperatively recognize the DENV replicative-form dsRNA to activate innate immunity

RIG-I like receptors (RLRs) are a family of cytosolic RNA sensors that sense RNA virus infection to activate innate immune response. It is generally believed that different RNA viruses are recognized by either RIG-I or MDA5, two important RLR members, depending on the nature of pathogen-associated molecular patterns (PAMPs) that are generated by RNA virus replication. Dengue virus (DENV) is an important RNA virus causing serious human diseases. Despite extensive investigations, the molecular basis of the DENV PAMP recognized by the host RLR has been poorly defined, and which RLR is involved in sensing the DENV PAMP remains controversial. Here, we demonstrated that the DENV infection-induced interferon response is dependent upon both RIG-I and MDA5, with RIG-I playing a predominant role. Next we purified the DENV PAMP RNA from the DENV-infected cells, and demonstrated that the purified DENV PAMP is viral full-length double-stranded RNA bearing 5ppp modifications, likely representing the viral replicative-form RNA. Finally, we confirmed the nature of the DENV PAMP by reconstituting the viral replicative-form RNA from in vitro synthesized DENV genomic RNA. In conclusion, our work not only defined the molecular basis of the RLR-PAMP interaction during DENV infection, but also revealed the previously underappreciated recognition of the distinct moiety of same PAMP by different RLRs in innate immunity against RNA viruses. ImportanceThe molecular interaction between PAMPs and RLRs plays a crucial role in innate immune response against the RNA virus infection. To our knowledge, the exact molecular basis of the DENV PAMPs and which RLR member (RIG-I or MDA5) is involved in this recognition remain controversial. In this study, we demonstrated that the DENV PAMP is likely DENV replicative-form RNA, a double-stranded RNA. RIG-I and MDA5 can sense different moieties of this DENV PAMP to active innate immune response. Our work not only clarified which RLR and what viral PAMP are involved in innate immune sensing of DENV infection, but also revealed previously underappreciated recognition of the same PAMPs by different RLRs in innate immunity against RNA viruses.

microbiology↗

bioRxiv2024

Manganese transporter SLC30A10 and iron transporters SLC40A1 and SLC11A2 impact dietary manganese absorption

SLC30A10 deficiency is a disease of severe manganese excess attributed to loss of SLC30A10-dependent manganese excretion via the gastrointestinal tract. Patients develop dystonia, cirrhosis, and polycythemia. They are treated with chelators but also respond to oral iron, suggesting that iron can outcompete manganese for absorption in this disease. Here we explore the latter observation. Intriguingly, manganese absorption is increased in Slc30a10-deficient mice despite manganese excess. Studies of multiple mouse models indicate that increased dietary manganese absorption reflects two processes: loss of manganese export from enterocytes into the gastrointestinal tract lumen by SLC30A10, and increased absorption of dietary manganese by iron transporters SLC11A2 (DMT1) and SLC40A1 (ferroportin). Our work demonstrates that aberrant absorption contributes prominently to SLC30A10 deficiency and expands our understanding of biological interactions between iron and manganese. Based on these results, we propose a reconsideration of the role of iron transporters in manganese homeostasis is warranted.

genetics↗

bioRxiv2023

Antibody Humanization via Protein Language Model and Neighbor Retrieval

Antibody (Ab), also known as immunoglobulin (Ig), is an essential macromolecule involved in human immune response and plays an increasingly vital role in drug discovery. However, the development of antibody drugs heavily relies on humanization of murine antibodies, which often necessitates multiple rounds of sequence optimizations through laborious experimental processes. In recent years, the remarkable capabilities of machine learning have revolutionized the field of natural sciences and have also demonstrated promising applications in the field of antibody humanization. Here, we present Protein-LAnguage-model-knN (PLAN), a machine learning model leveraging protein language model and information retrieval for improving humanization of antibodies. Further, we propose DE, a computed value shows a positive correlation with antigen-binding affinity. Our in silico experimental results demonstrate that 1) the PLAN-humanized sequences average humanness score reaches 0.592, improving over the best existing method by 44.7%; 2) a 63% overlap between the PLAN-proposed mutations and the mutations validated through wet lab experiments, which is 16.7% higher than the best existing result; 3) comparable antigen-binding affinity after DE guided back mutation.

bioinformatics↗

bioRxiv2022

DeepGWAS: Enhance GWAS Signals for Neuropsychiatric Disorders via Deep Neural Network

Genetic dissection of neuropsychiatric disorders can potentially reveal novel therapeutic targets. While genome-wide association studies (GWAS) have tremendously advanced our understanding, we approach a sample size bottleneck (i.e., the number of cases needed to identify >90% of all loci is impractical). Therefore, computationally enhancing GWAS on existing samples may be particularly valuable. Here, we describe DeepGWAS, a deep neural network-based method to enhance GWAS by integrating GWAS results with linkage disequilibrium and brain-related functional annotations. DeepGWAS enhanced schizophrenia (SCZ) loci by [~]3X when applied to the largest European GWAS, and 21.3% enhanced loci were validated by the latest multi-ancestry GWAS. Importantly, DeepGWAS models can be transferred to other neuropsychiatric disorders. Transferring SCZ-trained models to Alzheimers disease and major depressive disorder, we observed 1.3-17.6X detected loci compared to standard GWAS, among which 27-40% were validated by other GWAS studies. We anticipate DeepGWAS to be a powerful tool in GWAS studies.

genetics↗

bioRxiv2022

End-to-End deep structure generative model for protein design

AO_SCPLOWBSTRACTC_SCPLOWDesigning protein with desirable structure and functional properties is the pinnacle of computational protein design with unlimited potentials in the scientific community from therapeutic development to combating the global climate crisis. However, designing protein macromolecules at scale remains challenging due to hard-to-realize structures and low sequence design success rate. Recently, many generative models are proposed for protein design but they come with many limitations. Here, we present a VAE-based universal protein structure generative model that can model proteins in a large fold space and generate high-quality realistic 3-dimensional protein structures. We illustrate how our model can enable robust and efficient protein design pipelines with generated conformational decoys that bridge the gap in designing structure conforming sequences. Specifically, sequences generated from our design pipeline outperform native fixed backbone design in 856 out of the 1,016 tested targets(84.3%) through AF2 validation. We also demonstrate our models design capability and structural pre-training potential by structurally inpainting the complementarity-determining regions(CDRs) in a set of monoclonal antibodies and achieving superior performance compared to existing methods.

bioinformatics↗

bioRxiv2022

A Deep SE(3)-Equivariant Model for Learning Inverse Protein Folding

In this work, we establish a framework to tackle the inverse protein design problem; the task of predicting a proteins primary sequence given its backbone conformation. To this end, we develop a generative SE(3)-equivariant model which significantly improves upon existing autoregressive methods. Conditioned on backbone structure, and trained with our novel partial masking scheme and side-chain conformation loss, we achieve state-of-the-art native sequence recovery on structurally independent CASP13, CASP14, CATH4.2, and TS50 test sets. On top of accurately recovering native sequences, we demonstrate that our model captures functional aspects of the underlying protein by accurately predicting the effects of point mutations through testing on Deep Mutational Scanning datasets. We further verify the efficacy of our approach by comparing with recently proposed inverse protein folding methods and by rigorous ablation studies.

bioinformatics↗

bioRxiv2021

Genome-wide H3K9 Acetylation Level Increases with Age-Dependent Senescence of Flag Leaf in Rice (Oryza sativa)

Flag leaf senescence is an important biological process that drives the remobilization of nutrients to the growing organs of rice. Leaf senescence is controlled by genetic information via gene expression and epigenetic modification, but the precise mechanism is as of yet unclear. Here, we analyzed genome-wide acetylated lysine residue 9 of histone H3 (H3K9ac) enrichment by chromatin immunoprecipitation-sequencing (ChIP-seq) and examined its association with transcriptomes by RNA-seq during flag leaf aging in rice (Oryza sativa). We found that genome-wide H3K9 acetylation levels increased with age-dependent senescence in rice flag leaf, and there was a positive correlation between the density and breadth of H3K9ac and gene expression and transcript elongation. A set of 1,249 up-regulated, differentially expressed genes (DEGs) and 996 down-regulated DEGs showing a strong relationship between temporal changes in gene expression and gain/loss of H3K9ac was observed during rice flag leaf aging. We produced a landscape of H3K9 acetylation-modified gene expression targets that includes known senescence-associated genes, metabolism-related genes, as well as miRNA biosynthesis-related genes. Our findings reveal a complex regulatory network of metabolism- and senescence-related pathways mediated by H3K9ac and also elucidate patterns of H3K9ac-mediated regulation of gene expression during flag leaf aging in rice. Significance statementGenome-wide H3K9 acetylation levels increased with age-dependent senescence in rice flag leaf, and positively correlation the density and breadth of H3K9ac with transcript elongation and expression. Identified numerous H3K9 acetylation-modified gene expression targets reveal a complex regulatory network and metabolism-mediated senescence network that are associated with H3K9ac during leaf aging in rice.

plant biology↗

bioRxiv2021

Accurate Protein Function Prediction via Graph Attention Networks with Predicted Structure Information

Experimental protein function annotation does not scale with the fast-growing sequence databases. Only a tiny fraction (<0.1%) of protein sequences in UniProtKB has experimentally determined functional annotations. Computational methods may predict protein function in a high-throughput way, but its accuracy is not very satisfactory. Based upon recent breakthroughs in protein structure prediction and protein language models, we develop GAT-GO, a graph attention network (GAT) method that may substantially improve protein function prediction by leveraging predicted inter-residue contact graphs and protein sequence embedding. Our experimental results show that GAT-GO greatly outperforms the latest sequence- and structure-based deep learning methods. On the PDB-mmseqs testset where the train and test proteins share <15% sequence identity, GAT-GO yields Fmax(maximum F-score) 0.508, 0.416, 0.501, and AUPRC(area under the precision-recall curve) 0.427, 0.253, 0.411 for the MFO, BPO, CCO ontology domains, respectively, much better than homology-based method BLAST (Fmax 0.117,0.121,0.207 and AUPRC 0.120, 0.120, 0.163). On the PDB-cdhit testset where the training and test proteins share higher sequence identity, GAT-GO obtains Fmax 0.637, 0.501, 0.542 for the MFO, BPO, CCO ontology domains, respectively, and AUPRC 0.662, 0.384, 0.481, significantly exceeding the just-published graph convolution method DeepFRI, which has Fmax 0.542, 0.425, 0.424 and AUPRC 0.313, 0.159, 0.193.

bioinformatics↗

bioRxiv2021

Heterogeneity and molecular programming of progenitors for motor neurons and oligodendrocytes

The pMN domain is a restricted domain in the ventral spinal cords, defined by the expression of olig2 gene. The fate determination of pMN progenitors is highly temporally and spatially regulated, with motor neurons and oligodendrocyte progenitor cells (OPCs) developing sequentially. Insight into the heterogeneity and molecular programs of pMN progenitors is currently lacking. With the zebrafish model, we identified multiple states of neural progenitors using single-cell sequencing: proliferating progenitors, common progenitors for both motor neurons and OPCs, and restricted precursors for either motor neurons or OPCs. We found specific molecular programs for neural progenitor fate transition, and manipulations of representative genes in the motor neuron or OPC lineage confirmed their critical role in cell fate determination. The transcription factor NPAS3 is necessary for the development of the OPC lineage and can interact with many known genes associated with schizophrenia. Deciphering progenitor heterogeneity and molecular mechanisms for these transitions will elucidate the formation of complex neuron-glia networks in the central nervous system during development, and understand the basis of neurodevelopmental disorders.

neuroscience↗

bioRxiv2021

Predicting Epigenomic Functions of Genetic Variants in the Context of Neurodevelopment via Deep Transfer Learning

Decoding the regulatory effects of non-coding variants is a key challenge in understanding the mechanisms of gene regulation as well as the genetics of common diseases. Recently, deep learning models have been introduced to predict genome-wide epigenomic profiles and effects of DNA variants, in various cellular contexts, but they were often trained in cell lines or bulk tissues that may not be related to phenotypes of interest. This is particularly a challenge for neuropsychiatric disorders, since the most relevant cell and tissue types are often missing in the training data of such models. To address this issue, we introduce a deep transfer learning framework termed MetaChrom that takes advantage of both a reference dataset - an extensive compendium of publicly available epigenomic data, and epigenomic profiles of cell types related to specific phenotypes of interest. We trained and evaluated our model on a comprehensive set of epigenomic profiles from fetal and adult brain, and cellular models representing early neurodevelopment. MetaChrom predicts these epigenomic features with much higher accuracy than previous methods, and than models without the use of reference epigenomic data for transfer learning. Using experimentally determined regulatory variants from iPS cell-derived neurons, we show that MetaChrom predicts functional variants more accurately than existing non-coding variant scoring tools. By combining genome-wide association study (GWAS) data with MetaChrom predictions, we prioritized 31 SNPs for Schizophrenia (SCZ). These candidate SNPs suggest potential risk genes of SCZ and the biological contexts where they act. In summary, MetaChrom is a general transfer learning framework that can be applied to the study of regulatory functions of DNA sequences and variants in any disease-related cell or tissue types. The software tool is available at https://github.com/bl-2633/MetaChrom and a prediction web server is accessible at https://metachrom.ttic.edu/.

bioinformatics↗

bioRxiv2020

Reversing Transcriptome-Wide Association Studies to improve expression Quantitative Trait Loci associations

1Transcriptome-Wide Association Studies discover SNP effects mediated by gene expression through a two-stage process: a typically small reference panel is used to infer SNP-expression effects, and then these are applied to discover associations between imputed expression and phenotypes. We investigate whether the accuracy of SNP-expression and expression-phenotype associations can be increased by performing inference on both the reference panel and independent GWAS cohorts simultaneously. We develop EMBER (Estimation of Mediated Binary Effects in Regression) to re-estimate these effects using a liability threshold model with an adjustment to variance components accounting for imputed expression from GWAS data. In simulated data with only gene-mediated effects, EMBER more than doubles the performance of SNP-expression linear regression, increasing mean r2 from 0.3 to 0.65 with a gene-mediated variance explained of 0.01. EMBER also improves estimation accuracy when the fraction of cis-SNP variance mediated by genes is as low as 30%. We apply EMBER to genotype and gene expression data in schizophrenia by combining 512 samples from the CommonMind Consortium and 56,081 samples from the Psychiatric Genomic Consortium. We evaluate performance of EMBER in 36 genes suggested by TWAS by concordance of inferred effects with effects reported independently for frontal cortex expression. Applying the EMBER framework to a baseline linear regression model increases performance in 26 out of 36 genes (sign test p-value .0020) with an increase in mean r2 from 0.200 to 0.235.

genomics↗