Biology subjects

Walter, T.

Publications and source records attributed to Walter, T..

30 records · Page 2Linked to original sources

bioRxiv2022

Neural network for the prediction of treatment response in triple negative breast cancer

AO_SCPLOWBSTRACTC_SCPLOWThe automatic analysis of stained histological sections is becoming increasingly popular. Deep Learning is today the method of choice for the computational analysis of such data, and has shown spectacular results for large datasets for a large variety of cancer types and prediction tasks. On the other hand, many scientific questions relate to small, highly specific cohorts. Such cohorts pose serious challenges for Deep Learning, typically trained on large datasets. In this article, we propose a modification of the standard nested cross-validation procedure for hyper-parameter tuning and model selection, dedicated to the analysis of small cohorts. We also propose a new architecture for the particularly challenging question of treatment prediction, and apply this workflow to the prediction of response to neoadjuvant chemotherapy for Triple Negative Breast Cancer.

bioinformatics↗

bioRxiv2022

Automatic grading of cervical biopsies by combining full and self-supervision

In computational pathology, predictive models from Whole Slide Images (WSI) mostly rely on Multiple Instance Learning (MIL), where the WSI are represented as a bag of tiles, each of which is encoded by a Neural Network (NN). Slide-level predictions are then achieved by building models on the agglomeration of these tile encodings. The tile encoding strategy thus plays a key role for such models. Current approaches include the use of encodings trained on unrelated data sources, full supervision or self-supervision. While self-supervised learning (SSL) exploits unlabeled data, it often requires large computational resources to train. On the other end of the spectrum, fully-supervised methods make use of valuable prior knowledge about the data but involve a costly amount of expert time. This paper proposes a framework to reconcile SSL and full supervision, showing that a combination of both provides efficient encodings, both in terms of performance and in terms of biological interpretability. On a recently organized challenge on grading Cervical Biopsies, we show that our mixed supervision scheme reaches high performance (weighted accuracy (WA): 0.945), outperforming both SSL (WA: 0.927) and transfer learning from ImageNet (WA: 0.877). We further shed light upon the internal representations that trigger classification results, providing a method to reveal relevant phenotypic patterns for grading cervical biopsies. We expect that the combination of full and self-supervision is an interesting strategy for many tasks in computational pathology and will be widely adopted by the field.

cancer biology↗

bioRxiv2021

Deep Learning identifies new morphological patterns of Homologous Recombination Deficiency in luminal breast cancers from whole slide images.

Homologous Recombination DNA-repair deficiency (HRD) is a well-recognized marker of platinum-salt and PARP inhibitor chemotherapies in ovarian and breast cancers (BC). Causing high genomic instability, HRD is currently determined by BRCA1/2 sequencing or by genomic signatures, but its morphological manifestation is not well understood. Deep Learning (DL) is a powerful machine learning technique that has been recently shown to be capable of predicting genomic signatures from stained tissue slides. However, DL is known to be sensitive to dataset biases and lacks interpretability. Here, we present and evaluate a strategy to control for biases in retrospective cohorts. We train a deep-learning model to predict the HRD in a controlled cohort with unprecedented accuracy (AUC: 0.86) and we develop a new visualization technique that allows for automatic extraction of new morphological features related to HRD. We analyze in detail the extracted morphological patterns that open new hypotheses on the phenotypic impact of HRD.

cancer biology↗

bioRxiv2021

FISH-quant v2: a scalable and modular analysis tool for smFISH image analysis

Regulation of RNA abundance and localization is a key step in gene expression control. Single-molecule RNA fluorescence in-situ hybridization (smFISH) is a widely used single-cell-single-molecule imaging technique enabling a quantitative understanding of gene expression and its regulatory mechanisms. Recent progress in experimental techniques provides larger data-sets, requiring adequate tools for data analysis and exploration. Here, we present FISH-quant v2, a highly modular analysis tool accessible both for non-experts and experts, which we validated and applied on large-scale smFISH image datasets. Our package allows the user to detect isolated and clustered mRNA spots, segment nuclei and cells, quantify RNA localization patterns and visualize these results at the single-cell level.

bioinformatics↗

bioRxiv2021

Antibody evasion by the Brazilian P.1 strain of SARS-CoV-2

Terminating the SARS-CoV-2 pandemic relies upon pan-global vaccination. Current vaccines elicit neutralizing antibody responses to the virus spike derived from early isolates. However, new strains have emerged with multiple mutations: P.1 from Brazil, B.1.351 from South Africa and B.1.1.7 from the UK (12, 10 and 9 changes in the spike respectively). All have mutations in the ACE2 binding site with P.1 and B.1.351 having a virtually identical triplet: E484K, K417N/T and N501Y, which we show confer similar increased affinity for ACE2. We show that, surprisingly, P.1 is significantly less resistant to naturally acquired or vaccine induced antibody responses than B.1.351 suggesting that changes outside the RBD impact neutralisation. Monoclonal antibody 222 neutralises all three variants despite interacting with two of the ACE2 binding site mutations, we explain this through structural analysis and use the 222 light chain to largely restore neutralization potency to a major class of public antibodies.

microbiology↗

bioRxiv2020

The kinesin KIF1C transports APC-dependent mRNAs to cell protrusions

RNA localization and local translation are important for numerous cellular functions. In mammals, a class of mRNAs localize to cytoplasmic protrusions in an APC-dependent manner, with roles during cell migration. Here, we investigated this localization mechanism. We found that the KIF1C motor interacts with APC-dependent mRNAs and is required for their localization. Live cell imaging revealed rapid, active transport of single mRNAs over long distances that requires both microtubules and KIF1C. Two color imaging directly showed single mRNAs transported by single KIF1C motors, with the 3UTR being sufficient to trigger KIF1C-dependent RNA transport and localization. Moreover, KIF1C remained associated with peripheral, multimeric RNA clusters and was required for their formation. These results reveal an RNA transport pathway in mammalian cells, in which the KIF1C motor has a dual role both in transporting RNAs and in promoting their clustering within cytoplasmic protrusions. Interestingly, KIF1C also transports its own mRNA suggesting a possible feedback loop acting at the level of mRNA transport.

cell biology↗

bioRxiv2020

Neural network fast-classifies biological images using features selected after their random-forests-importance to power smart microscopy.

Artificial intelligence is nowadays used for cell detection and classification in optical microscopy, during post-acquisition analysis. The microscopes are now fully automated and next expected to be smart, to make acquisition decisions based on the images. It calls for analysing them on the fly. Biology further imposes training on a reduced dataset due to cost and time to prepare the samples and have the datasets annotated by experts. We propose here a real-time image processing, compliant with these specifications by balancing accurate detection and execution performance. We characterised the images using a generic, high-dimensional feature extractor. We then classified the images using machine learning for the sake of understanding the contribution of each feature in decision and execution time. We found that the non-linear-classifier random forests outperformed Fishers linear discriminant. More importantly, the most discriminant and time-consuming features could be excluded without any significant loss in accuracy, offering a substantial gain in execution time. It suggests a feature-group redundancy likely related to the biology of the observed cells. We offer a method to select fast and discriminant features. In our assay, a 79.6 {+/-} 2.4 % accurate classification of a cell took 68.7 {+/-} 3.5 ms (mean {+/-} SD, 5-fold cross-validation nested in 10 bootstrap repeats), corresponding to 14 cells per second, dispatched into 8 phases of the cell cycle using 12 feature-groups and operating a consumer market ARM-based embedded system. Interestingly, a simple neural network offered similar performances paving the way to faster training and classification, using parallel execution on a general-purpose graphic processing unit. Finally, this strategy is also usable for deep neural networks paving the way to optimising these algorithms for smart microscopy.

bioinformatics↗

bioRxiv2020

TRex, a fast multi-animal tracking system with markerless identification, 2D body posture estimation and visual field reconstruction

Automated visual tracking of animals is rapidly becoming an indispensable tool for the study of behavior. It offers a quantitative methodology by which organisms sensing and decision-making can be studied in a wide range of ecological contexts. Despite this, existing solutions tend to be challenging to deploy in practice, especially when considering long and/or high-resolution video-streams. Here, we present TRex, a fast and easy-to-use solution for tracking a large number of individuals simultaneously using background-subtraction with real-time (60Hz) tracking performance for up to approximately 256 individuals and estimates 2D visual-fields, outlines, and head/rear of bilateral animals, both in open and closed-loop contexts. Additionally, TRex offers highly-accurate, deep-learning-based visual identification of up to approximately 100 unmarked individuals, where it is between 2.5-46.7 times faster, and requires 2-10 times less memory, than comparable software (with relative performance increasing for more organisms/longer videos) and provides interactive data-exploration within an intuitive, platform-independent graphical user-interface.

animal behavior and cognition↗

bioRxiv2020

Both HIV Infection and Conditional Tat Expression Decrease Prepulse Inhibition with Further Impairment by Methamphetamine History in Humans

HIV infection and methamphetamine (METH) use are highly comorbid and represent a significant public health problem. Both conditions are known to negatively impact a variety of brain functions. One brain function that may be affected by HIV and METH use is sensorimotor gating, an automatic, pre-conscious filtering of sensory information that is thought to contribute to higher order cognitive processes. Sensorimotor gating is often measured using prepulse inhibition (PPI), a paradigm that can be conducted in both humans and animals, thereby enabling cross-species translational studies. While previous studies suggest HIV and METH may individually impair PPI, little research has been conducted on the effects of combined HIV and METH on PPI. The goal of this cross-species study was to determine the effects of METH on PPI in the inducible Tat (iTat) mouse model of HIV and in people with HIV. PPI was measured in the iTat mouse model before, during, and after chronic METH treatment and after Tat induction. Chronic METH treatment decreased PPI in male but not female mice. PPI normalized with cessation of METH. Inducing Tat expression decreased PPI in male but not in female mice. No interactions between chronic METH treatment and Tat expression were observed in mice. In humans, HIV was associated with decreased PPI in both men and women. Furthermore, PPI was lowest in people with HIV who also had a history of METH dependence. Overall, these results suggest HIV and METH may additively impair early information processing in humans, potentially affecting downstream cognitive function. HIGHLIGHTSO_LIHIV decreased PPI in men and women C_LIO_LIPPI was most decreased in people with HIV and a history of METH dependence C_LIO_LIChronic METH treatment decreased PPI in male but not female mice C_LIO_LITat expression decreased PPI in male but not female mice C_LIO_LIChronic METH treatment and Tat expression did not interact to affect PPI in mice C_LI

neuroscience↗

bioRxiv2020

A localization screen reveals translation factories and widespread co-translational RNA targeting

Local translation allows a spatial control of gene expression. Here, we performed a dual protein/mRNA localization screen, using smFISH on 523 human cell lines expressing GFP-tagged genes. A total of 32 mRNAs displayed specific cytoplasmic localizations, and we observed local translation at unexpected locations, including cytoplasmic protrusions, cell edges, endosomes, Golgi, the nuclear envelope and centrosomes, the latter being cell cycle dependent. Quantitation of mRNA distribution and automatic pattern classification revealed a high degree of localization heterogeneity between cells. Surprisingly, mRNA localization frequently required ongoing translation, indicating widespread co-translational RNA targeting. Interestingly, while P-body accumulation was frequent (15 mRNAs), four mRNAs accumulated in foci that were distinct structures. These foci lacked the mature protein, but nascent polypeptide imaging showed that they were specialized translation factories. For {beta}-catenin, foci formation was regulated by Wnt, relied on APC-dependent polysome aggregation, and led to nascent protein degradation. Thus, translation factories uniquely regulate nascent protein metabolism and create a fine granular compartmentalization of translation.

cell biology↗

bioRxiv2019

Domain-invariant features for mechanism of action prediction in a multi-cell-line drug screen

High Content Screening is an important tool in drug discovery and characterisation. Often, high content drug screens are performed on one single cell line. Yet, a single cell line cannot be thought of as a perfect disease model. Many diseases feature an important molecular heterogeneity. Consequently, a drug may be effective against one molecular subtype of a disease, but less so against another. To characterise drugs with respect to their effect not only on one cell line but on a panel of cell lines is therefore a promising strategy to streamline the drug discovery process. The contribution of this paper is twofold. First, we investigate whether we can predict drug mechanism of action (MOA) at the molecular level without optimisation of the MOA classes to the screen specificities. To this end, we benchmark a set of algorithms within a conventional pipeline, and evaluate their MOA prediction performance according to a statistically rigorous framework. Second, we extend this conventional pipeline to the simultaneous analysis of multiple cell lines, each manifesting potentially different morphological baselines. For this, we propose multitask autoencoders, including a domain-adaptive model used to construct domain-invariant feature representations across cell lines. We apply these methods to a pilot screen of two triple negative breast cancer cell lines as models for two different molecular subtypes of the disease.

bioinformatics↗

bioRxiv2017

Kernel multitask regression for toxicogenetics

The development of high-throughput in vitro assays to study quantitatively the toxicity of chemical compounds on genetically characterized human-derived cell lines paves the way to predictive toxicogenetics, where one would be able to predict the toxicity of any particular compound on any particular individual. In this paper we present a machine learning-based approach for that purpose, kernel multitask regression (KMR), which combines chemical characterizations of molecular compounds with genetic and transcriptomic characterizations of cell lines to predict the toxicity of a given compound on a given cell line. We demonstrate the relevance of the method on the recent DREAM8 Toxicogenetics challenge, where it ranked among the best state-of-the-art models, and discuss the importance of choosing good descriptors for cell lines and chemicals.

bioinformatics↗