Biology subjects

Rapsomaniki, M. A.

Publications and source records attributed to Rapsomaniki, M. A..

6 recordsLinked to original sources

bioRxiv2023

Multiplexed tumor profiling with generative AI accelerates histopathology workflows and improves clinical predictions

Understanding the spatial heterogeneity of tumors and its links to disease is a cornerstone of cancer biology. Emerging spatial technologies offer unprecedented capabilities towards this goal, but several limitations hinder their clinical adoption. To date, histopathology workflows still heavily depend on hematoxylin & eosin (H&E) and serial immunohistochemistry (IHC) staining, a cumbersome and tissue-exhaustive process that yields unaligned tissue images. We propose the VirtualMultiplexer, a generative AI toolkit that translates real H&E images to matching IHC images for several markers based on contrastive learning. The VirtualMultiplexer learns from unpaired H&E and IHC images and introduces a novel multi-scale loss to ensure consistent and biologically reliable stainings. The virtually multiplexed images enabled training a Graph Transformer that simultaneously learns from the joint spatial distribution of several markers to predict clinically relevant endpoints. Our results indicate that the VirtualMultiplexer achieves rapid, robust and precise generation of virtually multiplexed imaging datasets of high staining quality that are indistinguishable from the real ones. We successfully employed transfer learning to generate realistic virtual stainings across tissue scales, patient cohorts, and cancer types with no need for model fine-tuning. Crucially, the generated images are not only realistic but also clinically relevant, as they greatly improved the prediction of different clinical endpoints across patient cohorts and cancer types, speeding up histopathology workflows and accelerating spatial biology.

bioinformatics↗

bioRxiv2022

ChromFormer: A transformer-based model for 3Dgenome structure prediction

Recent research has shown that the three-dimensional (3D) genome structure is strongly linked to cell function. Modeling the 3D genome structure can not only elucidate vital biological processes, but also reveal structural disruptions linked to disease. In the absence of experimental techniques able to determine 3D chromatin structure, this task is achieved computationally by exploiting chromatin interaction frequencies as measured by high-throughput chromosome conformation capture (Hi-C) data. However, existing methods are unsupervised, and limited by underlying assumptions. In this work, we present a novel framework for 3D chromatin structure prediction from Hi-C data. The framework consists of, a novel synthetic data generation module that simulates realistic structures and corresponding Hi-C matrices, and CO_SCPLOWHROMC_SCPLOWFO_SCPLOWORMERC_SCPLOW, a transformer-based model to predict 3D chromatin structures from standalone Hi-C data, while providing local structural-level confidence estimates. Our solution outperforms existing methods when tested on unseen synthetic data, and achieves comparable results on experimental data for a full eukaryotic genome. The code, data, and models can be accessed at https://github.com/AI4SCR/ChromFormer.

bioinformatics↗

bioRxiv2022

MatchCLOT: Single-Cell Modality Matching with Contrastive Learning and Optimal Transport

Recent advances in single-cell technologies have enabled the simultaneous quantification of multiple biomolecules in the same cell, opening new avenues for understanding cellular complexity and heterogeneity. However, the resulting multimodal single-cell datasets present unique challenges arising from the high dimensionality of the data and the multiple sources of acquisition noise. In this work, we propose MO_SCPLOWATCHC_SCPLOWCLOT, a novel method for single-cell data integration based on ideas borrowed from contrastive learning, optimal transport, and transductive learning. In particular, we use contrastive learning to learn a common representation between two modalities and apply entropic optimal transport as an approximate maximum weight bipartite matching algorithm. Our model obtains state-of-the-art performance in the modality matching task from the NeurIPS 2021 multimodal single-cell data integration challenge, improving the previous best competition score by 28.9%. Our code can be accessed at https://github.com/AI4SCR/MatchCLOT.

bioinformatics↗

bioRxiv2021

Modeling the Three-Dimensional Chromatin Structure from Hi-C Data with Transfer Learning

Recent studies have revealed the importance of three-dimensional (3D) chromatin structure in the regulation of vital biological processes. Contrary to protein folding, no experimental procedure that can directly determine ground-truth 3D chromatin coordinates exists. Instead, chromatin conformation is studied implicitly using high-throughput chromosome conformation capture (Hi-C) methods that quantify the frequency of all pairwise chromatin contacts. Computational methods that infer the 3D chromatin structure from Hi-C data are thus unsupervised, and limited by the assumption that contact frequency determines Euclidean distance. Inspired by recent developments in deep learning, in this work we explore the idea of transfer learning to address the crucial lack of ground-truth data for 3D chromatin structure inference. We present a novel method, Transfer learning Encoder for CHromatin 3D structure prediction (TECH-3D) that combines transfer learning with creative data generation procedures to reconstruct chromatin structure. Our work outperforms previous deep learning attempts for chromatin structure inference and exhibits similar results as state-of-the-art algorithms on many tests, without making any assumptions on the relationship between contact frequencies and Euclidean distances. Above all, TECH-3D presents a highly creative and novel approach, paving the way for future deep learning models.

bioinformatics↗

bioRxiv2020

In silico analysis of DNA re-replication across a complete genome reveals cell-to-cell heterogeneity and genome plasticity

DNA replication is a complex and remarkably robust process: despite its inherent uncertainty, manifested through stochastic replication timing at a single-cell level, multiple control mechanisms ensure its accurate and timely completion across a population. Disruptions in these mechanisms lead to DNA re-replication, closely connected to genomic instability and oncogenesis. We present a stochastic hybrid model of DNA re-replication that accurately portrays the interplay between discrete dynamics, continuous dynamics, and uncertainty. Using experimental data on the fission yeast genome, model simulations show how different regions respond to re-replication, and permit insight into the key mechanisms affecting re-replication dynamics. Simulated and experimental population-level profiles exhibit good correlation along the genome, which is robust to model parameters, validating our approach. At a single-cell level, copy numbers of individual loci are affected by intrinsic properties of each locus, in cis effects from adjoining loci and in trans effects from distant loci. In silico analysis and single-cell imaging reveal that cell-to-cell heterogeneity is inherent in re-replication and can lead to a plethora of genotypic variations. Our thorough in silico analysis of DNA re-replication across a complete genome reveals that heterogeneity at the single cell level and robustness at the population level are emerging and co-existing principles of DNA re-replication. Our results indicate that re-replication can promote genome plasticity by generating many diverse genotypes within a population, potentially offering an evolutionary advantage in cells with aberrations in replication control mechanisms.

systems biology↗

bioRxiv2019

A stochastic hybrid model for DNA replication incorporating protein mobility dynamics

DNA replication, the basis of genetic information maintenance, is a remarkably robust yet highly stochastic process. We present a computational model that incorporates experimental genome structures and protein mobility dynamics to mechanistically describe the stochastic foundations of DNA replication. Analysis of about 300,000 in silico profiles for fission yeast indicates that the number of firing factors is rate-limiting and dominates completion time. Incorporating probabilistic activation and binding, a full-genome duplication was achieved with at least 300 firing factors, with the only assumption that factors get recycled upon replication fork collision. Spatial patterns of replication timing were reproduced only when firing factors were explicitly activated proximally to the spindle pole body. Independent in vivo experiments corroborate that the spindle pole body acts as a replication activator, driving origin firing. Our model provides a framework to realistically simulate full-genome DNA replication and investigate the effects of nuclear architecture.

bioinformatics↗