Biology subjects

GUO, Y.

Publications and source records attributed to GUO, Y..

5 recordsLinked to original sources

bioRxiv2026

Interpretable Deep Learning of Endogenous Fluorophore Concentration and Entropy for Lung Tissue Classification

Fluorophore concentration measures and local histogram entropy describe different aspects of endogenous fluorophore maps. We investigate whether their combination improves discrimination between normal and cancerous lung tissue and which features contribute to predictions. Chemometric fluorescence microscopy (CFM) provides concentration maps of tryptophan, elastin, reduced nicotinamide adenine dinucleotide (NADH), and flavin adenine dinucleotide (FAD). A neural network based on generalized additive models with structured interactions (GAMI-Net) is used to compare concentration-only, entropy-only, and combined representations through explicit main effects and pairwise interactions. The mean areas under the receiver operating characteristic curve (ROC AUCs) across five-fold cross-validation were $95.9\% \pm 0.7\%$, $96.4\% \pm 1.6\%$, and $98.4\% \pm 0.7\%$, respectively; variability is reported as standard deviation (SD). Their aggregate differences are consistent with the hypothesis of complementary predictive information. FAD entropy, NADH concentration, and NADH entropy were the leading main effects in the combined model. The learned log-odds contributions include a U-shaped elastin-entropy relationship and nonmonotonic tryptophan relationships. By exposing how concentration and entropy enter predictions, this interpretable framework turns label-free fluorescence classification into specific, testable hypotheses about tissue metabolism and extracellular-matrix organization, motivating specimen-grouped and external validation.

biophysics↗

bioRxiv2025

Plant Face: Machine learning decodes genetic, environmental and developmental imprints in leaf appearance

Leaf appearance is a crucial plant phenotype. However, traditional methods for extracting this information are inefficient, limiting its full utilization. Deep learning based on convolutional neural networks (CNNs) enables us to capture previously inaccessible information from images. In this study, we made the surprising discovery that the leaf appearance of each individual plant is unique. Using deep learning, leaves from one plant could be efficiently distinguished from those of another plant of the same species and cultivar. We term this phenomenon the "Plant Face" and suggest the potential to develop a "plant face recognition system," analogous to human facial recognition. We also applied similar methods to study the relationship between leaflet appearance and their position on compound leaves, leaf bilateral symmetry, and differences in leaves from twining stems with different chirality. These results collectively indicate that plant genetic characteristics, growth conditions, and developmental features can be stored within their appearance. With appropriate decoding, leaf appearance is poised to play an increasingly important role in phenomics. SignificanceThe saying "no two leaves in the world are identical" holds philosophical significance, as such variation encompasses considerable contingency and randomness. Here, we assert that no two trees have identical leaves; meaning that even for plants of the same species and cultivar, the leaf morphology of each individual plant is distinct at the population level, even though single leaves may overlap in appearance. Genetic, environmental, and developmental information is recorded in some manner within the phenotypic appearance of leaves. With advancements in computational technologies like artificial intelligence, this information can now be decoded. HighlightsO_LIThe leaves of each individual plant are statistically unique. C_LIO_LIThe relationship between leaflet appearances in compound leaves hints at their developmental patterns. C_LIO_LILeaves are not necessarily bilaterally symmetric in a statistical sense. C_LIO_LILeaves from stems with different chirality (twining direction) exhibit distinct appearances. C_LI

plant biology↗

bioRxiv2024

Chemically Induced Nuclear Pore Complex Protein Degradation via E3 Ligase TRIM21

Despite the exciting progress of the bifunctional degrader molecules, also known as proteolysis-targeting chimeras (PROTACs), the rapidly expanding field is still significantly hampered by the lacking of available E3 ligase ligands. Our research bridges this gap by uncovering a series of small-molecule ligands to the E3 ligase TRIM21 through DNA-Encoded Library (DEL) technology. We confirmed their interaction with TRIM21 using crystallography and demonstrated their anti-proliferative effects across various cancer cell types. Furthermore, proteomic studies identified that the mRNA Export Factor GLE1, and the Nuclear Pore Complex Protein NUP155, were significantly down-regulated upon TRIM21 ligand treatment. This degradation required TRIM21 and was ubiquitin-proteasome-dependent. More specifically, NUP155 was the primary target for the TRIM21 ligands, while GLE1 was considered a passenger target upon the initial degradation of NUP155. Using immunofluorescence techniques, we further demonstrated that the degradation of GLE1 and NUP155 proteins impaired the integrity of the nuclear envelope, leading to cell death. Highlighted by this research, a novel mode of action has been discovered for the TRIM21 E3 ligase ligand, acting as a monovalent degrader that triggers de novo interaction with functional complex proteins and induces their degradation.

cell biology↗

bioRxiv2021

Systematic Chromosome Rearrangement Induced by CRISPR-Cas9 Reshapes the Genome and Transcriptome of Human Cells

Chromosome rearrangement plays important roles in development, carcinogenesis and evolution. However, its mechanism and subsequent effects are not fully understood. At present, large-scale chromosome rearrangement has been performed in the simple eukaryote, wine yeast, but the relative research in mammalian cells remains at the level of individual chromosome rearrangement due to technical limitations. In this study, we used CRISPR-Cas9 to target the highly repetitive human endogenous retrotransposons, LINE-1 (L1) and Alu, resulting in a large number of DNA double-strand breaks in the chromosomes. While this operation killed the majority of the cells, we eventually obtained live cell groups. Karyotype analysis and genome re-sequencing proved that we have achieved systematic chromosome rearrangement (SCR) in human cells. The copy number variations (CNVs) of the SCR genomes showed typical patterns that observed in tumor genomes. For example, the most frequent deleted region Chr9p21 containing p15 and p16 tumor suppressor, and the amplified region Chr8q24 containing MYC in tumors were all identified in both SCR cells. The ATAC-seq and RNA-seq further revealed that the epigenetic and transcriptomic landscapes were deeply reshaped by the SCR. Gene expressions related to p53 pathway, DNA repair, cell cycle and apoptosis were greatly altered to facilitate the cell survival under the severe stress induced by the large-scale chromosomal breaks. In addition, we found that the cells acquired CRISPR-Cas9 resistance after SCR by interfering with the Cas9 mRNA. Our study provided a new application of CRISPR-Cas9 and a practical approach for SCR in complex mammalian genomes. HighlightsRepetitive retroelements with large copy numbers were targeted using CRISPR-Cas9 in human cells. Cells survived after their chromosomal DNAs were heavily cleaved and systematic chromosome rearrangement was achieved. Systematic chromosome rearrangement reshaped the genome and transcriptome of the cells.

genomics↗

bioRxiv2016

DNAlogo: a smart mini application for generating DNA sequence logos

Sequence logos are frequently used for presenting consensus sequences and motifs of nucleic acids and proteins. WebLogo of UC Berkeley is the most popular tool for creating sequence logos, whereas, it is often restricted by the internet speed in some developing countries and no graphic interface for its stand-alone version. Here, the author generated an application, DNAlogo, using VB.net, which runs in Windows system. DNAlogo is small and convenient. It creates both bitmap and vector map. Beside the classic sequence logo function, DNAlogo introduced compensated logo to generate consensus sequences considering different GC contents of different genomes. DNAlogo provides a simple way for researchers without programming knowledge to create DNA sequence logos.

bioinformatics↗