A5085123886- Single-cell and spatial transcriptomics
- Gene expression and cancer classification
- Retinal Development and Disorders
- Statistical Methods and Inference
- Neural dynamics and brain function
- Visual perception and processing mechanisms
- Cancer Genomics and Diagnostics
- Extracellular vesicles in disease
- Artificial Intelligence in Healthcare and Education
- Photoreceptor and optogenetics research
- Gene Regulatory Network Analysis
- Neuroscience and Neuropharmacology Research
- Cell Image Analysis Techniques
- Topic Modeling
Hertie Institute for Clinical Brain Research
2023-2024
University of Tübingen
2020-2024
Bernstein Center for Computational Neuroscience Tübingen
2021-2023
Tübingen AI Center
2023
Standard preprocessing of single-cell RNA-seq UMI data includes normalization by sequencing depth to remove this technical variability, and nonlinear transformation stabilize the variance across genes with different expression levels. Instead, two recent papers propose use statistical count models for these tasks: Hafemeister Satija (Genome Biol 20:296, 2019) recommend using Pearson residuals from negative binomial regression, while Townes et al. 20:295, fitting a generalized PCA model....
Recent large language models (LLMs) can generate and revise text with human-level performance, have been widely commercialized in systems like ChatGPT. These come clear limitations: they produce inaccurate information, reinforce existing biases, be easily misused. Yet, many scientists using them to assist their scholarly writing. How wide-spread is LLM usage the academic literature currently? To answer this question, we use an unbiased, large-scale approach, free from any assumptions on...
A recent paper in
A recent paper claimed that t -SNE and UMAP embeddings of single-cell datasets are “specious” fail to capture true biological structure. The authors argued such as arbitrary misleading forcing the data into an elephant shape. Here we show this conclusion was based on inadequate limited metrics embedding quality. More appropriate quantifying neighborhood class preservation reveal in room: while do not preserve high-dimensional distances, they can nevertheless provide biologically relevant information.
Abstract Background Standard preprocessing of single-cell RNA-seq UMI data includes normalization by sequencing depth to remove this technical variability, and nonlinear transformation stabilize the variance across genes with different expression levels. Instead, two recent papers propose use statistical count models for these tasks: Hafemeister & Satija [1] recommend using Pearson residuals from negative binomial regression, while Townes et al. [2] fitting a generalized PCA model. Here,...
Identifying cell types and understanding their functional properties is crucial for unraveling the mechanisms underlying perception cognition. In retina, can be identified by carefully selected stimuli, but this requires expert domain knowledge biases procedure towards previously known types. visual cortex, it still unknown what exist how to identify them. Thus, unbiased identification of in retina new approaches are needed. Here we propose an optimization-based clustering approach using...
Abstract Recent work employed Pearson residuals from Poisson or negative binomial models to normalize UMI data. To extend this approach non-UMI data, we model the additional amplification step with a compound distribution: assume that sequenced RNA molecules follow distribution, and are then replicated following an distribution. We show how leads residuals, which yield meaningful gene selection embeddings of Smart-seq2 datasets. Further, suggest distributions across several sequencing...