The Human dataset provides a comprehensive collection of drug-target interactions specific to human proteins, aimed at facilitating research in drug discovery and bioinformatics. This dataset includes a diverse range of human proteins as drug targets, along with associated drug molecules and their respective interaction labels. The data consists of molecular descriptors of drugs, protein sequences, and experimentally validated interactions sourced from various biological databases.
Create Graph：we collected asymmetric drug-drug interaction (DDI) entries from version 5.1.12 of DrugBank, released on March 14, 2024. After a thorough double-check, we removed drugs with incorrect SMILES strings or those that could not be represented by Morgan fingerprints . This filtering resulted in a dataset containing 1,752 drugs and 508,512 asymmetric interactions. Subsequently, we organized the DDI entries into a directed interaction network, where directed edges represent the asymmetric interactions between drugs.
The 'AirScript' dataset consists of surface electromyography (sEMG) signals obtained while writing the uppercase English alphabets (A–Z) in free space. The Delsys Trigno device was used to record forearm muscle activity from 16 subjects. Every subject performs two trials for each letter, thus resulting in 52 samples per subject. sEMG signals obtained from all subjects were stored at a 2000 Hz sampling rate for high temporal resolution. The dataset consists of raw sEMG signals that are stored in subject-specific folders and saved as `.npy` files.
This paper presents data on the life of 1000 Bedlington Terriers. Observations were conducted from 2015 to 2021. The oldest pet in 2021 was 17 years old, the youngest was 6.
The table shows the correlation between the life expectancy of Bedlington Terriers with copper toxicosis and healthy ones.
On average, the life expectancy of sick dogs is reduced by 44%, however, about 12% of dogs are susceptible to the disease. 40% of sick dogs die within 3 years, the rest within 5 years, extremely rare are individuals living more than 5 years.
This dataset comprises extensive multi-modal data related to the experimental study of ultrasonically excited pulsating fluid jets used for bone cement removal. Conducted at the Institute of Geonics, Ostrava, Czech Republic, the study explores the effect of varying standoff distances on erosion profiles, under controlled parameters including a fixed nozzle diameter, sonotrode frequency, supply pressure, and robot arm velocity. The dataset includes numerical data representing ablation profiles, captured as a large CSV file, and audio recordings captured using a high-resolution microphone.
We provide a curated time series video dataset of 10,876 frames from 37 neonates, of which some were healthy and some had asphyxia of varying severity. We obtained video recordings using standardized RGB cameras positioned to capture the skin region of the neonates while minimizing interference with standard clinical care and maintaining consistent video quality. This study has the following technical specifications: a frame rate of 30 frames per second, a resolution of 1920×1080 pixels, and a color depth of 24-bit RGB.
Precise prediction of potential drug-disease associations (DDAs) is essential for enhancing treatment strategies and expediting drug development. However, current methods often rely on single-modal data and fail to effectively integrate multimodal information when representing node attributes.
As various modalities of genomic data are accumulating, methods to integrate across multi-omics datasets are becoming important. Error-correcting output codes (ECOC) is an ensemble learning strategy for solving a multiclass problem thru a decoding process that aggregates the predictions of multiple classifiers. Thus, it lends itself naturally to aggregating predictions across multiple views as well. We applied the ECOC to multi-view learning to see if this strategy can enhance classifier performance as compared to traditional techniques.
In structure-based drug design (SBDD), a major challenge is generating high-affinity 3D ligand molecules that can effectively bind to specific protein targets, which requires accurately capturing complex protein-ligand interactions. Although existing diffusion models have demonstrated potential in molecular generation tasks, they often struggle with accurately capturing the complex interactions between proteins and ligands. To address this problem, we propose MSIDiff, a multi-stage interaction-aware diffusion model for protein-specific molecular generation.
Osteoarthritis (OA) is a prevalent degenerative joint disease,particularly affecting the knees. Early and accurate detection of OA and its severity, often graded using the Kellgren-Lawrence (KL) scale, is crucial for timely intervention and management. This study explores the application of deep learning techniques to automatically detect OA and assign KL grades from knee X-ray images. We propose a novel deep learning architecture that effectively extracts relevant features from X-ray images and classifies them into different KL grades.
