Biology subjects

Platt, R. J.

Publications and source records attributed to Platt, R. J..

6 recordsLinked to original sources

bioRxiv2026

Acquired epithelial Wnt7b secretion establishes autocrine WNT dependency in BRAF mutant colorectal cancer

Colorectal cancer (CRC) subtypes differ fundamentally in their reliance on WNT signaling, yet the sources and regulation of WNT ligands as well as the subtype-specific requirements for ligand-driven activation remain poorly defined. Here, we generated genetically-engineered organoid models of Apc- and Braf-mutant CRC to systematically dissect ligand dependency across distinct genetic backgrounds and CRC subtypes. Through genetic perturbation and pharmacological inhibition of WNT secretion, we find that BRAF-driven organoids critically depend on autocrine WNT ligand production for survival and proliferation. In contrast, APC-mutant organoids remain viable but undergo discernible transcriptional changes upon ligand withdrawal. We identify Wnt7b as a non-redundant epithelial WNT ligand required to sustain beta-catenin-dependent signaling specifically in BRAF-mutant CRC. Chromatin accessibility and transcriptional profiling reveal that Wnt7b activation occurs early during transformation and coincides with cancer-specific opening of regulatory elements at the Wnt7b locus. Integrative regulatory analyses nominate the transcription factor RFX7 as a candidate upstream regulator of epithelial Wnt7b expression. Consistent with these findings, analysis of human CRC specimens demonstrates WNT7B expression in a subset of BRAF-mutant tumors and reveals an association with poor patient survival. Together, these results redefine ligand dependency in colorectal cancer and uncover autocrine WNT7B signaling as a mechanistically defined vulnerability in aggressive BRAF-mutant disease.

molecular biology↗

bioRxiv2026

Reliable single-cell perturbations explain and improve model performance

Predicting single-cell transcriptional responses to perturbations is central to building the virtual cell, yet recent benchmarks show that simple baseline methods often outperform complex models, and model comparisons depend on the evaluation metric. Most studies assume that preprocessed RNA sequencing data are reliable ground truth for both training and evaluation. Here, we test this assumption by measuring the reliability of perturbations and their alignment with shared perturbation responses, classifying each perturbation as specific, shared, or unreliable. Among 7,170 perturbations from 29 datasets, 65% are unreliable, 11% shared, and 24% specific. Applying these quality labels to published benchmarks shows that model comparisons depend on perturbation quality. Training with reliable perturbations alone matches or outperforms full-data performance while using 55% of all training perturbations. Our framework also enables prospective experimental design: for most perturbations, a 28-cell pilot experiment accurately predicts how many cells a full screen needs to be reliable.

bioinformatics↗

bioRxiv2025

Perturbation-aware representation learning for in vivo genetic screens

CRISPR-based genetic perturbation screens paired with single-cell transcriptomic readouts (Perturb-seq) offer a powerful tool for interrogating biological systems. Yet the resulting datasets are heterogeneous--particularly in vivo--and currently used cell-level perturbation labels reflect only CRISPR guide RNA exposure rather than perturbation state; further, many perturbations have a minimal effect on gene expression. For perturbations that do alter the transcriptomic state of cells, intracellular guide RNA abundance exhibits a dose-response association with perturbation efficacy. We combine (i) per-perturbation, expression-only classifiers trained with non-negative negative-unlabeled (nnNU) risk to yield calibrated scores reflecting the perturbation state of single cells and (ii) a monotone guide abundance prior to yield a per-cell pseudo-posterior that supports both assignment of perturbation probability and selection of affected gene features. To obtain a low-dimensional representation that allows for the accurate reconstruction of gene-level marginals for counterfactual decoding, we train an autoencoder with a quantile-hurdle reconstruction loss and feature-weighted emphasis on perturbation-affected genes. The result is a perturbation-aware latent embedding amenable to downstream trajectory modeling (e.g., optimal transport or flow matching) and a principled probability of perturbation for each non-control cell derived jointly from its guide counts and transcriptome.

bioinformatics↗

bioRxiv2024

An oncoembryology approach uncovers SoxC-driven regulation of colon development and cancer

Reactivated embryonic programs are associated with cancer progression, yet their role and regulatory mechanisms remain poorly understood. In this study, we introduce oncoembryology, an approach that systematically compares embryonic and cancerous tissues to identify shared molecular programs and assess their functional relevance in disease. Applying this strategy to colorectal cancer, we identified SoxC transcription factors (Sox4, Sox11, Sox12) as critical regulators of both embryonic development and tumorigenesis. SoxC transcription factors regulate diverse downstream targets, including Tead2, Mdk, and Klf4, thereby regulating crucial steps of colon development. Abrogating SoxC function in murine models reduced tumor growth and prevented liver metastasis. Concordantly, a SoxC-driven oncoembryonic gene signature correlated with poor survival in colorectal cancer patients, underscoring the therapeutic potential of targeting SoxC-regulated pathways in cancer treatment.

cancer biology↗

bioRxiv2023

Developmental molecular controls over arealization of descending cortical motor pathways

Layer 5 extratelencephalic (ET) neurons are a main class of neocortical projection neurons that predominate in the motor cortex and send their axon to the pons and spinal cord, and collaterals to the thalamus and multiple deep subcerebral structures1-3. Precise connectivity of ET neurons is critical for fine motor control; they are central to loss of function upon spinal cord injury and specifically degenerate in select neurodegenerative disorders4, 5. ET neurons consist of several types of cells with distinct laminar and areal locations, molecular identities, connectivities, and functions6, 7. Within layer 5 of the cortex, two cardinal subtypes of ET neurons have been identified: "ETlower" neurons, which express Slco2a1 and project to distal targets including the spinal cord, "ETupper" neurons, which express Nprs1 or Hpgd and project more proximally to the pons and thalamus6. Despite their critical function, how these neuronal subtypes emerge during development and acquire their area-specific distributions remains unaddressed. Here, using combinations of anatomical labeling, MAPseq mapping8, and single-nucleus transcriptomics across developing cortical areas, we reveal that these two subtypes of ET neurons are present at birth along opposite antero-posterior cortical gradients. We first characterize area-specific developmental axonal dynamics of ETlower and ETupper neurons and find that the latter can emerge by pruning of subsets of ETlower neurons. We next identify area- and ET neuron type-specific developmental transcriptional programs to identify key target genes in vivo. Finally, we reprogram ET neuron area-specific connectivity from motor to visual by postnatal in vivo combinatorial knockout of three key type-specific transcription factors. Together, these findings delineate the functional transcriptional programs controlling ET neuron diversity across cortical areas and provide a molecular blueprint to investigate and direct the developmental emergence of corticospinal motor control.

neuroscience↗

bioRxiv2021

MiR-137 and miR-122, two outer subventricular zone-enriched non-coding RNAs, regulate basal progenitor expansion and neuronal differentiation

Cortical expansion in the primate brain relies on the presence and the spatial enlargement of multiple germinal zones during development and on a prolonged developmental period. In contrast to other mammals, which have two cortical germinal zones, the ventricular zone (VZ) and subventricular zone (SVZ), gyrencephalic species display an additional germinal zone, the outer subventricular zone (OSVZ), which role is to increase the number and types of neurons generated during corticogenesis. How the OSVZ emerged during evolution is poorly understood but recent studies suggest a role for non-coding RNAs, which allow tight regulations of transcriptional programs in time and space during development (Dehay et al. 2015; Arcila et al., 2014). Here, using in vivo functional genetics, single-cell RNA sequencing, live imaging and electrophysiology to assess progenitor and neuronal properties in mice, we identify two ferret and human OSVZ-enriched microRNAs (miR), miR-137 and miR-122, which regulate key cellular features associated with cortical expansion. MiR-137 promotes basal progenitor self-replication and superficial layer neuron fate, while miR-122 slows down neuronal differentiation pace. Together, these findings support a cell-type specific role for miR-mediated transcriptional regulation in cortical expansion.

evolutionary biology↗