Biology subjects

Baek, I.

Publications and source records attributed to Baek, I..

8 recordsLinked to original sources

bioRxiv2025

From minutes to bounds: A probabilistic UV-C control and a shape-only morphological fingerprint for postharvest Colletotrichum

Postharvest losses in high-value horticultural crops such as cacao and coffee are often driven by Colletotrichum spp. and other latent fruit pathogens. Ultraviolet-C (UV-C) is increasingly deployed as a chemical-free postharvest technology, yet prescriptions remain framed in minutes rather than in probabilistic guarantees of disease control and safety. We present a dual framework that (1) establishes conservative confidence bounds on survival and (2) validates a "shape-only" morphological fingerprint. This approach addresses the biological complexity of host-pathogen interactions by quantifying isolate-specific heterogeneity, rather than averaging it away. We utilized a large Colletotrichum dataset (n = 5,363) from cacao and coffee, spanning diverse treatments including UV-C, UV-B, and sonication. First, focusing on the Coffee UV-C cohort ([~]10 min), we quantified this heterogeneity; the most conservative Clopper-Pearson upper 95% bound on survival reached 1.000, highlighting partial survival events (e.g., isolate P24-88) under otherwise high-efficacy conditions. This probabilistic framework captures the "tail-risk" of biological resilience instead of assuming complete kill. Second, we trained machine learning models on the full dataset using only geometric features (e.g., aspect ratio, asymmetry), explicitly excluding all primary size metrics. Serving as a rapid physiological indicator of UV-induced stress, these "shape-only" models successfully predicted pathogen host-origin (Accuracy {approx} 0.93) and post-treatment survival (R{superscript 2} {approx} 0.74). The signals ability to generalize across UV-B and sonication confirms that geometry, not just growth reduction, carries a robust and transferable physiological stress signature. This work provides a device-agnostic, probabilistic control platform, replacing time-based heuristics with quantitative guarantees and a generalizable, shape-based diagnostic. HighlightsO_LIA conservative upper 95 % confidence bound (on survival = 1.000) was observed for the Coffee UV-C ([~]10 min) cohort, revealing strong isolate-specific heterogeneity rather than a universal kill guarantee C_LIO_LIIsolate-specific heterogeneity (e.g., localized P24-88 survival) was quantified rather than averaged away. C_LIO_LIA size-free morphological fingerprint predicted host-origin (Accuracy {approx} 0.93) and survival (R{superscript 2} {approx} 0.74). C_LIO_LIThe shape-only signal generalized across diverse stressors, including UV-C, UV-B, and sonication. C_LIO_LIA probabilistic, device-agnostic control framework replaces traditional time-based heuristics. C_LI

microbiology↗

bioRxiv2025

Decoding pathogen ecological memory from chemical-stress phenotypes

Can standardized chemical stress reveal a reproducible, taxonomy-agnostic stress-response fingerprint that is predictive of sample-source labels (crop-of-isolation) in fungal isolates? Six coffee{square}associated Colletotrichum isolates were profiled using four phenolic{square}branched compounds and compared them with a previously characterized cacao panel. Quantitative morphology, hyperspectral imaging (HSI), and supervised machine learning (ML) yielded panel-specific fingerprints under uniform, isotropic in vitro conditions. Circularity, a measure of edge symmetry, was the most informative morphological feature, and ML classified the crop-of-isolation label (coffee vs cacao, in this panel) with 86.7% accuracy in within-panel cross-validation. HSI detected dose-dependent spectral shifts in a targeted subset of isolates and compounds, including changes near 1930 nm in the short-wave infrared, a moisture-sensitive region that warrants robustness checks (e.g., band masking or preprocessing sensitivity) prior to biochemical attribution. Multi{square}locus phylogeny showed the coffee isolates are polyphyletic, so the predictive signal should be interpreted conservatively as a taxonomy-agnostic phenotype fingerprint associated with crop background in this mixed-lineage panel, acknowledging that crop labels are partially confounded with phylogenetic structure. We propose a "chemical priors" framework as a working hypothesis, in which long-term environmental exposure may imprint stress-response pathways that become legible under simple, standardized probes. This integrative workflow supports scalable screening of eco-friendly antifungals and sensor-driven decision support for high-throughput phenotype-based screening workflows. HighlightsO_LITaxonomy-agnostic workflow integrates hyperspectral imaging and morphology. C_LIO_LIMachine learning predicts crop-source labels with 86.7% accuracy in mixed lineages. C_LIO_LISystem exhibits robustness, maintaining >86% accuracy even after feature ablation. C_LIO_LINon-linear ML captures structure missed by linear stats (1.7% variance). C_LIO_LIEnables rapid, sensor-driven antifungal screening without prior DNA sequencing. C_LI

microbiology↗

bioRxiv2025

An Eligibility-Aware Pipeline for Robust ITS Diagnostics in Fungi: A Cacao Case Study with Generalizable Rules

Accurate fungal ITS diagnostics rely on public sequence archives, but heterogeneous record lengths, especially frequent truncation before the LSU/28S segment, cause naive in silico benchmarking to conflate primer performance with database incompleteness. To resolve this persistent "denominator error," we present an open and fully reproducible "eligibility-aware" framework. Our pipeline first establishes eligibility by confirming both primer sites are present before applying bench-realistic performance rules, including a strict penalty for 3-terminal mismatches. It further provides mechanistic insights by analyzing binding-site conservation and uses a rarefaction-based approach to guide efficient quality control as databases grow. We demonstrate the frameworks utility using the cacao pathosystem, a context where rapid differentiation of the fungal pathogen Moniliophthora from symptomatically similar oomycetes is critical. The result is a robust, field-ready diagnostic decision tree operable under a single touchdown (TD) PCR program. By providing a transparent and barcode-agnostic template, our eligibility-aware approach offers a significant methodological advance for designing and validating molecular assays in mycology and beyond.

microbiology↗

bioRxiv2025

Unveiling differential responses to UVB (305 nm) and UVC (275 nm) in cacao-infecting Colletotrichum gloeosporioides and Pestalotiopsis sp.

Sustainable control of microbial pathogens requires alternatives to chemicals, but physical methods like Ultraviolet-C (UVC) show variable efficacy linked to poorly understood pathogen-specific responses. Here, we investigate differential UVB/UVC responses in plant pathogenic fungi (Colletotrichum gloeosporioides, Pestalotiopsis sp.). Using hyperspectral imaging and machine learning, we dissect the physiological underpinnings of UV sensitivity. UVC proves more potent than UVB, with Pestalotiopsis sp. showing significantly higher resistance than C. gloeosporioides isolates. Crucially, hyperspectral signatures correlated with resistance, revealing UVC-induced photopigment changes, biochemical disruption, and oxidative stress markers in sensitive isolates, contrasting with minimal perturbation in the resistant isolate. Machine learning accurately decoded these complex phenotypes for classification. This understanding enabled enhanced inactivation via optimized pulsed UVC and synergistic sonication. We link distinct physiological states, non-invasively detected via hyperspectral imaging, to fungal UV resistance, demonstrating how integrating advanced phenotyping and machine learning provides a mechanistic basis for optimizing physical pathogen controls.

microbiology↗

bioRxiv2025

Genome-wide association mapping and predictive modeling of wet bean mass in a diverse cacao collection

Improving cacao yield, a key objective in post-domestication crop improvement, remains a primary goal for breeders, but progress is often hindered by the confounding effects of population structure. To overcome this, we analyzed 346 diverse cacao accessions using an ML-based association mapping framework (with and without population structure adjustment) and a phenotype-only ML prediction of yield. By correcting for population structure, our Bootstrap Forest-based GWAS revealed association signals that showed consistent enrichment for ribosome and protein-synthesis functions, and a recurrent subset of SNPs with high importance appeared across multiple yield components, including pod index and seed number. In parallel, a Neural Network model was utilized to identify cotyledon mass and length as the most powerful predictors for total wet bean mass (R{superscript 2} = 0.715 by repeated five-fold cross-validation), suggesting a practical, low-cost screening proxy for breeding). Collectively, this study delivers a robust genetic framework and a novel predictive tool to accelerate the development of high-yielding cacao varieties through the early identification of elite clones.

genomics↗

bioRxiv2023

Hypothalamic neuronal activation in primates drives naturalistic goal-directed eating behavior

Eating addiction is the primary cause of modern obesity. Although the causal role of the lateral hypothalamic area (LHA) for eating is demonstrated in rodents, there is no evidence in primates regarding naturalistic eating behaviors. We investigated the role of LHA GABAergic (LHAGABA) neurons in eating by chemogenetics in three macaques. LHAGABA neuron activation significantly increased naturalistic goal-directed behaviors and food motivation, which was specific for palatable food. PET and MRS validated the chemogenetic activation. Rs-fMRI result revealed that functional connectivity (FC) between the LHA and frontal areas was increased, while the FC between the frontal cortices was decreased after the LHAGABA neuron activation. Thus, our study elucidates the role of LHAGABA neurons in eating and obesity therapeutics for primates and humans.

neuroscience↗

bioRxiv2021

Single molecule studies reveal branched pathways for activator-dependent pre-initiation complex assembly

RNA polymerase II (Pol II) transcription reconstituted from purified factors suggests pre-initiation complexes (PICs) can assemble by sequential incorporation of factors at the TATA box. However, these basal transcription reactions are generally independent of activators and co-activators. To study PIC assembly under more realistic conditions, we used single-molecule microscopy to visualize factor dynamics during activator-dependent reactions in nuclear extracts. Surprisingly, Pol II, TFIIF, and TFIIE can pre-assemble on enhancer-bound activators before loading into PICs, and multiple Pol II complexes can bind simultaneously to create a localized cluster. Unlike TFIIF and TFIIE, TFIIH binding is singular and dependent on the basal promoter. Activator-tethered factors exhibit dwell times on the order of seconds. In contrast, PICs can persist on the order of minutes in the absence of nucleotide triphosphates, although TFIIE remains unexpectedly dynamic even after TFIIH incorporation. Our kinetic measurements lead to a new branched model for activator-dependent PIC assembly. HIGHLIGHTSSingle molecule microscopy reveals unexpected dynamics of RNA Pol II and GTFs Multiple Pol IIs cluster on UAS/enhancer-bound activators before binding the core promoter Pol II, TFIIF, and TFIIE, but not TFIIH, can pre-assemble at the UAS/enhancer Activators increase the rates of Pol II and GTF association with DNA eTOC BlurbSingle-molecule microscopy experiments by Baek et al. show that RNA polymerase II and basal transcription factors TFIIF and TFIIE preassemble on UAS/enhancer-bound activators, poised for loading into initiation complexes with TFIIH at the core promoter. Transcription activators kinetically enhance factor recruitment, creating a localized cluster of polymerases at the UAS/enhancer.

molecular biology↗

bioRxiv2020

Dynamics of RNA polymerase II and elongation factor Spt4/5 recruitment during activator-dependent transcription

In eukaryotes, RNA polymerase II (RNApII) transcribes messenger RNA from template DNA. Decades of experiments have identified the proteins needed for transcription activation, initiation complex assembly, and productive elongation. However, the dynamics of recruitment of these proteins to transcription complexes, and of the transitions between these steps, are poorly understood. We used multi-wavelength single-molecule fluorescence microscopy to directly image and quantitate these dynamics in a budding yeast nuclear extract that reconstitutes activator-dependent transcription in vitro. A strong activator (Gal4-VP16) greatly stimulated reversible binding of individual RNApII molecules to template DNA, with no detectable involvement of RNApII-containing condensates. Binding of labeled elongation factor Spt4/5 to DNA typically followed RNApII binding, was NTP-dependent, and was correlated with association of mRNA-binding protein Hek2, demonstrating specificity of Spt4/5 binding to elongation complexes. Quantitative kinetic modeling shows that only a fraction of RNApII binding events are productive and implies a rate-limiting step, probably associated with recruitment of general transcription factors, needed to assemble a transcription-competent pre-initiation complex at the promoter. Spt4/5 association with transcription complexes was slowly reversible, with DNA-bound RNApII molecules sometimes binding and releasing Spt4/5 multiple times. The average Spt4/5 residence time was of similar magnitude to the time required to transcribe an average length yeast gene. These dynamics suggest that a single Spt4/5 molecule remains associated during a typical transcription event, yet can dissociate from RNApII to allow disassembly of abnormally long-lived (i.e., stalled) elongation complexes. Significance StatementThe synthesis of a eukaryotic messenger RNA molecule involves the association of RNA polymerase and dozens of accessory proteins on DNA. We used differently colored fluorescent dyes to tag DNA, RNA polymerase II, and the elongation factor Spt4/5 in yeast nuclear extract, and then observed the assembly and dynamics of individual molecules of the proteins with single DNA molecules by microscopy. The observations quantitatively define an overall pathway by which transcription complexes form and evolve in response to an activator protein. They suggest how molecular complex dynamics may be tuned to optimize efficient RNA production.

biochemistry↗