Search arXiv⌕ Search

arXiv subjects

Carl Herrmann

Publications and source records attributed to Carl Herrmann.

11 recordsLinked to original sources
arXiv2026

On the Promises and Limits of Multi-omics Integration for Deconvolution: The HADACA3 Benchmark

Understanding the cellular composition of complex tissues, such as tumors, is a key challenge in biology and medicine. A common approach, known as deconvolution, aims to estimate the cellular composition from bulk molecular measurements. With the growing availability of multiple types of molecular data, it is often assumed that combining data sources should improve deconvolution performance. Here, we present HADACA3, a community-driven benchmark designed to evaluate this assumption. We conducted a four-day collaborative competition followed by a large-scale computational benchmark, testing more than 250,000 analysis pipelines across nine datasets with matched DNA methylation (DNAm) and RNA profiles, representing a wide range of biological and experimental conditions. Our framework jointly evaluates the impact of preprocessing, feature selection, modeling, and integration strategies. We find that DNAm alone achieves the highest median performance across datasets, making it the most stable and reliable single-modality approach. However, multi-omics integration strategies can regularly achieve higher top performance in specific datasets and pipeline configurations. Among the tested strategies, late integration based on error-weighted averaging provides a strong and reliable baseline, while non-linear early integration methods, such as optimal transport, show promising results on real biological datasets. Overall, our results show that multi-omics integration does not systematically improve average performance over DNAm alone, but can improve best-case performance in specific settings. This highlights a trade-off between robustness and peak performance, and emphasizes the importance of aligning integration strategies with the statistical properties of the data. All data, code, and evaluation tools are publicly available to support reproducible research and future method development.

q-bio.QM↗
arXiv2026

Skipping the Zeros in Diffusion Models for Sparse Data Generation

Diffusion models (DMs) excel on dense continuous data, but are not designed for sparse continuous data. They do not model exact zeros that represent the deliberate absence of a signal. As a result, they erase sparsity patterns and perform unnecessary computation on mostly zero entries. With Sparsity-Exploiting Diffusion (SED), we model only non-zero values, preserving sparsity. SED delivers computational savings while maintaining or improving generation quality by skipping zeros during training and inference. Across physics and biology benchmarks, SED matches or surpasses conventional DMs and domain-specific baselines, while vision experiments provide intuitive insights into the limitations of dense DMs and the benefits of SED.

cs.LG↗
arXiv2025

Sparse Data Diffusion for Scientific Simulations in Biology and Physics

Sparse data is fundamental to scientific simulations in biology and physics, from single-cell gene expression to particle calorimetry, where exact zeros encode physical absence rather than weak signal. However, existing diffusion models lack the physical rigor to faithfully represent this sparsity. This work introduces Sparse Data Diffusion (SDD), a generative method that explicitly models exact zeros via Sparsity Bits, unifying efficient ML generation with physically grounded sparsity handling. Empirical validation in particle physics and single-cell biology demonstrates that SDD achieves higher fidelity than baseline methods in capturing sparse patterns critical for scientific analysis, advancing scalable and physically faithful simulation.

cs.LG↗
arXiv2005

Ab initio identification of putative human transcription factor binding sites by comparative genomics

We discuss a simple and powerful approach for the ab initio identification of cis-regulatory motifs involved in transcriptional regulation. The method we present integrates several elements: human-mouse comparison, statistical analysis of genomic sequences and the concept of coregulation. We apply it to a complete scan of the human genome. By using the catalogue of conserved upstream sequences collected in the CORG database we construct sets of genes sharing the same overrepresented motif (short DNA sequence) in their upstream regions both in human and in mouse. We perform this construction for all possible motifs from 5 to 8 nucleotides in length and then filter the resulting sets looking for two types of evidence of coregulation: first, we analyze the Gene Ontology annotation of the genes in the set, searching for statistically significant common annotations; second, we analyze the expression profiles of the genes in the set as measured by microarray experiments, searching for evidence of coexpression. The sets which pass one or both filters are conjectured to contain a significant fraction of coregulated genes, and the upstream motifs characterizing the sets are thus good candidates to be the binding sites of the TF's involved in such regulation. In this way we find various known motifs and also some new candidate binding sites.

q-bio.GN↗
arXiv2003

Connectivity Distribution of Spatial Networks

We study spatial networks constructed by randomly placing nodes on a manifold and joining two nodes with an edge whenever their distance is less than a certain cutoff. We derive the general expression for the connectivity distribution of such networks as a functional of the distribution of the nodes. We show that for regular spatial densities, the corresponding spatial network has a connectivity distribution decreasing faster than an exponential. In contrast, we also show that scale-free networks with a power law decreasing connectivity distribution are obtained when a certain information measure of the node distribution (integral of higher powers of the distribution) diverges. We illustrate our results on a simple example for which we present simulation results. Finally, we speculate on the role played by the limiting case P(k)=1/k which appears empirically to be relevant to spatial networks of biological origin such as the ones constructed from gene expression data.

cond-mat.dis-nn↗
arXiv2001

General Matter Coupled N=4 Gauged Supergravity in Five Dimensions

We construct the general form of matter coupled N=4 gauged supergravity in five dimensions. Depending on the structure of the gauge group, these theories are found to involve vector and/or tensor multiplets. When self-dual tensor fields are present, they must be charged under a one-dimensional Abelian group and cannot transform non-trivially under any other part of the gauge group. A short analysis of the possible ground states of the different types of theories is given. It is found that AdS ground states are only possible when the gauge group is a direct product of a one-dimensional Abelian group and a semi-simple group. In the purely Abelian, as well as in the purely semi-simple gauging, at most Minkowski ground states are possible. The existence of such Minkowski ground states could be proven in the compact Abelian case.

hep-th↗
arXiv2000

N=4 Supergravity with Antisymmetric Tensor in Central Charge Superspace

A concise geometrical formulation of N=4 supergravity containing an antisymmetric tensor gauge field is given in central charge superspace: graviphotons are identified in the super-vielbein on the same footing as the vierbein and the Rarita-Schwinger fields. As a consequence of superspace soldering, Chern-Simons terms in the fieldstrength of the antisymmetric tensor arise as an intrinsic property of superspace with central charge coordinates.

hep-th↗
arXiv2000

Domain walls in five dimensional supergravity with non-trivial hypermultiplets

We study BPS domain wall solutions of 5-dimensional N=2 supergravity where isometries of the hypermultiplet geometry have been gauged. We derive the corresponding supersymmetric flow equations and define an appropriate c-function. As an example we discuss a domain wall solution of Freedman, Gubser, Pilch and Warner which is related to a RG-flow in a dual superconformal field theory.

hep-th↗
arXiv1999

Compactification of Type IIB String Theory on Calabi-Yau Threefolds

We study compactifications of type IIB supergravity on Calabi-Yau threefolds. The resulting low energy effective Lagrangian is displayed in the large volume limit and its symmetry properties - with specific emphasis on the SL(2,Z) - are discussed. The explicit map to type IIA string theory compactified on a mirror Calabi-Yau is derived. We argue that strong coupling effects on the worldsheet break the SL(2,Z).

hep-th↗
arXiv1997

The N=2 vector-tensor multiplet, central charge superspace, and Chern-Simons couplings

We present a new, alternative interpretation of the vector-tensor multiplet as a 2-form in central charge superspace. This approach provides a geometric description of the (non-trivial) central charge transformations ab initio and is naturally generalized to include couplings of Chern-Simons forms to the antisymmetric tensor gauge field, giving rise to a N=2 supersymmetric version of the Green-Schwarz anomaly cancellation mechanism.

hep-th↗