arXiv research

ChemCPA predicts cellular responses to novel drugs using transfer learning.

Paper compresses RNNs using HT decomposition for better performance.

New insights into why neural networks generalize well.

Let S be a compact, connected, orientable surface of positive genus. Let HT(S) be the Hatcher-Thurston complex of S. We prove that Aut(HT(S)) is isomorphic to the extended mapping class group of S modulo its center.

Machine learning improves RNA secondary structure prediction.

Designing RNA molecules has garnered recent interest in medicine, synthetic biology, biotechnology and bioinformatics since many functional RNA molecules were shown to be involved in regulatory processes for transcription, epigenetics and translation. Since an RNA's function depends on its structural properties, the RN…

RNA-binding proteins (RBPs) play crucial roles in many biological processes, e.g. gene regulation. Computational identification of RBP binding sites on RNAs are urgently needed. In particular, RBPs bind to RNAs by recognizing sequence motifs. Thus, fast locating those motifs on RNA sequences is crucial and time-efficie…

Generalized quandle polynomial used for stuquandles, stuck links, and RNA folding.

Deep learning predicts RNA degradation from crowdsourced data.

A new $K$-means method HT $K$-means uses $\ell_0$ penalty for sparsity.

Predicting RNA base distances using a large language model.

Solving the RNA inverse folding problem is a critical prerequisite to RNA design, an emerging field in bioengineering with a broad range of applications from reaction catalysis to cancer therapy. Although significant progress has been made in developing machine-based inverse RNA folding algorithms, current approaches s…

New tropical geometry connects handlebodies and outer space.

A deep learning model organizes RNA graphs to reveal folding patterns and properties.

New benchmarks for RNA 3D structure-function modeling.

In recent years, the advances in single-cell RNA-seq techniques have enabled us to perform large-scale transcriptomic profiling at single-cell resolution in a high-throughput manner. Unsupervised learning such as data clustering has become the central component to identify and characterize novel cell types and gene exp…

Let $R$ be a compact, connected, orientable surface of genus $g$ with $n$ boundary components with $g \geq 2$, $n \geq 0$. Let $\mathcal{N}(R)$ be the nonseparating curve graph, $\mathcal{C}(R)$ be the curve graph and $\mathcal{HT}(R)$ be the Hatcher-Thurston graph of $R$. We prove that if $λ: \mathcal{N}(R) \rightarro…

The Regularized Nonlinear Acceleration (RNA) algorithm is an acceleration method capable of improving the rate of convergence of many optimization schemes such as gradient descend, SAGA or SVRG. Until now, its analysis is limited to convex problems, but empirical observations shows that RNA may be extended to wider set…

New method detects RNA modifications without prior training, revealing novel sites.

Long non-coding RNAs (lncRNAs) are a class of non-coding RNAs which play a significant role in several biological processes. RNA-seq based transcriptome sequencing has been extensively used for identification of lncRNAs. However, accurate identification of lncRNAs in RNA-seq datasets is crucial for exploring their char…

Let $X$ be a compact, geodesically complete, locally CAT(0) space such that the universal cover admits a rank one axis. Assume $X$ is not homothetic to a metric graph with integer edge lengths. Let $P_t$ be the number of parallel classes of oriented closed geodesics of length $\le t$; then $\lim\limits_{t \to \infty} P…

New invariants for RNA foldings and stuck links defined.

Study uses knot theory to model RNA foldings, emphasizing both entanglement and intrachain interactions.

Improved GPLVM model for single-cell RNA-seq data.

The study counts geodesics on special manifolds without focusing points.

Given two or more Deep Neural Networks (DNNs) with the same or similar architectures, and trained on the same dataset, but trained with different solvers, parameters, hyper-parameters, regularization, etc., can we predict which DNN will have the best test accuracy, and can we do so without peeking at the test data? In …

HSSE framework embeds single-cell RNA-seq data at multiple scales.

Let $S_{1}$ and $S_{2}$ be connected orientable surfaces of genus $g_{1}, g_{2} \geq 3$, $n_{1},n_{2} \geq 0$ punctures, and empty boundary. Let also $\varphi: \mathcal{HT}(S_{1}) \rightarrow \mathcal{HT}(S_{2})$ be an edge-preserving alternating map between their Hatcher-Thurston graphs. We prove that $g_{1} \leq g_{2…

A novel text-independent speaker identification (SI) method is proposed. This method uses the Mel-frequency Cepstral coefficients (MFCCs) and the dynamic information among adjacent frames as feature sets to capture speaker's characteristics. In order to utilize dynamic information, we design super-MFCCs features by cas…

Non-coding RNA (ncRNA) are RNA sequences which don't code for a gene but instead carry important biological functions. The task of ncRNA classification consists in classifying a given ncRNA sequence into its family. While it has been shown that the graph structure of an ncRNA sequence folding is of great importance for…

Recent advances in high-throughput cDNA sequencing (RNA-Seq) technology have revolutionized transcriptome studies. A major motivation for RNA-Seq is to map the structure of expressed transcripts at nucleotide resolution. With accurate computational tools for transcript reconstruction, this technology may also become us…

In this paper, we propose an end-to-end deep learning model, called E2Efold, for RNA secondary structure prediction which can effectively take into account the inherent constraints in the problem. The key idea of E2Efold is to directly predict the RNA base-pairing matrix, and use an unrolled algorithm for constrained p…

New methods improve analysis of single cell RNA sequencing data.

Geometric model for a specific group in Artin groups.

RNA structures show that a significant portion of bases do not form hydrogen bonds.

The standard state-of-the-art backend for text-independent speaker recognizers that use i-vectors or x-vectors, is Gaussian PLDA (G-PLDA), assisted by a Gaussianization step involving length normalization. G-PLDA can be trained with both generative or discriminative methods. It has long been known that heavy-tailed PLD…

The paper benchmarks data stream classifiers for human activity recognition on connected devices.

Timely detection of Hardware Trojans (HTs) has become a major challenge for secure integrated circuits. We present a run-time methodology for HT detection that employs a multi-parameter statistical traffic modeling of the communication channel in a given System-on-Chip (SoC), named as SIMCom. The main idea is to model …

We study the secondary structure of RNA determined by Watson-Crick pairing without pseudo-knots using Milnor invariants of links. We focus on the first non-trivial invariant, which we call the Heisenberg invariant. The Heisenberg invariant, which is an integer, can be interpreted in terms of the Heisenberg group as wel…

TempBalance improves neural network training by balancing learning rates across layers.

Graph neural networks improve with edge similarity constraints in RNA structure analysis.

Symmetric CNNs improve sequential recommendation and protein structure prediction.

Flexible models cluster RNA sequencing data.

Regularized nonlinear acceleration (RNA) estimates the minimum of a function by post-processing iterates from an algorithm such as the gradient method. It can be seen as a regularized version of Anderson acceleration, a classical acceleration scheme from numerical analysis. The new scheme provably improves the rate of …

Treebolic space is an analog of the Sol geometry, namely, it is the horocylic product of the hyperbolic upper half plane H and the homogeneous tree T with degree p+1 > 2, the latter seen as a one-complex. Let h be the Busemann function of T with respect to a fixed boundary point. Then for real q > 1 and integer p > 1, …

In this work we propose a method to compute continuous embeddings for kmers from raw RNA-seq data, without the need for alignment to a reference genome. The approach uses an RNN to transform kmers of the RNA-seq reads into a 2 dimensional representation that is used to predict abundance of each kmer. We report that our…

Study analyzes neural network models to understand generalization performance.

A new model synthesizes population with fewer structural and sampling zeros.