Find research datasets worth reusing

FI_WD20K & SI_WD20K: Datasets For Hyper-Relational Inductive Link Prediction

The datasets correspond to the work presented in <a href="https://arxiv.org/abs/2107.04894">"Improving Inductive Link Prediction Using Hyper-Relational Facts"</a>.

Antioxidant capacity prediction of Nanoforms using supervised algorithms.

This dataset contains p-chem properties of Nanoforms and experimental conditions for the antioxidant capacity measured with the DPPH assay.

Data for "Predicting aggregate morphology of sequence-defined macromolecules with Recurrent Neural Networks"

These are the data associated with the paper, "Predicting aggregate morphology of sequence-defined macromolecules with Recurrent Neural Networks" (DOI 10.1039/D2SM00452F). Three of the directories contains subdirectories with `GSD` files dumped from HOOMD. The other contains pretrained RNN models as TorchScript binaries exported from PyTorch.

opencc-zeroMay 2022View details →

Scaling the leaf length-times-width equation to predict total leaf area of shoots

Background and Aims: An individual plant consists of different-sized shoots, each of which consists of different-sized leaves. To predict plant-level physiological responses from the responses of individual leaves, modelling this within-shoot leaf size variation is necessary. Within-plant leaf trait variation has been well investigated in canopy photosynthesis models but less so in plant allometry. Therefore, integration of these two different approaches is needed. Methods: We focused on an established leaf-level relationship that the area of an individual leaf lamina is proportional to the product of its length and width. The geometric interpretation of this equation is that different-sized leaf laminas from a single species share the same basic form. Based on this shared basic form, we synthesized a new length-times-width equation predicting total shoot leaf area from the collective dimensions of leaves that comprise a shoot. Furthermore, we showed that several previously established empirical relationships, including the allometric relationships between total shoot leaf area, maximum individual leaf length within the shoot and total leaf number of the shoot, can be unified under the same geometric argument. We tested the model predictions using five species, all of which have simple leaves, selected from diverse taxa (Magnoliids, monocots and eudicots) and from different growth forms (trees, erect herbs and rosette herbs). Key Results: For all five species, the length-times-width equation explained within-species variation of total leaf area of a shoot with high accuracy (R2 &gt; 0.994). These strong relationships existed despite leaf dimensions scaling very differently between species. We also found good support for all derived predictions from the model (R2 &gt; 0.85). Conclusions: Our model can be incorporated to improve previous models of allometry that do not consider within-shoot size variation of individual leaves, providing a cross-scale linkage between individual leaf-size variation and shoot-size variation.

opencc-zeroMay 2022View details →

Water availability dictates how plant traits predict demographic rates

A major goal in ecology is to make generalizable predictions of organism responses to environmental variation based on their traits. However, straightforward relationships between traits and fitness are rare and likely vary with environmental context. Characterizing how traits mediate demographic responses to the environment may enhance predictions of organism responses to global change. We synthesized 15 years of demographic data and species-level traits in a shortgrass steppe to determine whether the effects of leaf and root traits on growth and survival depend on seasonal water availability. We predicted that (1) species with drought-tolerant traits, such as lower leaf turgor loss point (TLP) and higher leaf and root dry matter content (LDMC and RDMC), would be more likely to survive and grow in drier years due to higher wilting resistance, (2) these traits would not predict fitness in wetter years, and (3) traits that more directly measure physiological mechanisms of water use such as TLP would best predict demographic responses. We found that graminoids with more negative TLP and higher LDMC and RDMC had higher survival rates in drier years. Forbs demonstrated similar yet more variable responses. Graminoids grew larger in wetter years, regardless of traits. However, in both wet and dry years, graminoids with more negative TLP and higher LDMC and RDMC grew larger than less negative TLP and low LDMC and RDMC species. Traits significantly mediated the impact of drought on survival, but not growth, suggesting survival could be a stronger driver of species' drought response in this system. TLP predicted survival in drier years, but easier-to-measure LDMC and RDMC were equal or better predictors. These results advance our understanding of the mechanisms by which drought drives population dynamics, and show that abiotic context determines how traits drive fitness.

Kinship matrices for deep learning for whole-genome predictions

Kinship matrices to be used for the deep learning model implementation used in Tensorflow/Keras for whole-genome predictions (see Github repo at https://github.com/filippob/paper_deep_learning_vs_gblup/)

Code and data repository for the role of topography, geotechnical layering, and attenuation on ground motion prediction

Data and scripts to reproduce research on the influence of topography, geotechnical layer, and attenuation on ground motion prediction in Salton Trough.

opencc-zeroJun 2022View details →

Beyond sex and aggression: Testosterone rapidly matches behavioral responses to social context and tries to predict the future

Although androgens are widely studied in the context of aggression, androgenic influences on prosocial behaviors have been less explored. We examined testosterone's (T) influence on prosocial and aggressive responses in a positively-valenced social context (interacting with a pairbond partner) and a negatively-valenced context (interacting with an intruder) in socially monogamous Mongolian gerbils. T increased and decreased prosocial responses in the same individuals towards a pairbond partner and an intruder, respectively, both within 30 minutes, but did not affect aggression. T also had persistent effects on prosocial behavior; males in which T initially increased prosocial responses towards a partner continued to exhibit elevated prosocial responses towards an intruder male days later until a second T injection rapidly eliminated those responses. Thus, T surges can rapidly match behavior to current social context, as well as prime animals for positive social interactions in the future. Neuroanatomically, T rapidly increased hypothalamic oxytocin, but not vasopressin, cellular responses during interactions with a partner. Together, our results indicate that T can facilitate and inhibit prosocial behaviors depending on social context, that it can influence prosocial responses across rapid and prolonged time scales, and that it affects oxytocin signaling mechanisms that could mediate its context-dependent behavioral influences.

opencc-zeroJun 2022View details →

Predicting foraging dive outcomes in chinstrap penguins using biologging and animal-borne cameras

Direct observation of foraging behavior is not always possible, especially for marine species that hunt below the surface. However, biologging and tracking devices in particular have provided very detailed information about how various species use their habitat. From these indirect observations, researchers have tried to infer foraging and prey catching events for a more accurate definition of these species' ecological niches. In this study, we deployed video cameras in addition to GPS and time-depth recorders on chinstrap penguins during the brood phase of the 2018-19 breeding season at various colonies on the Gourlay peninsula (South Orkney Islands). More than 57 hours of footage from 16 birds covering 770 dives were scrutinized by two independent observers. The outcome of each dive was classified as unsuccessful, individual krill encounter or krill swarm encounter. In addition, the number of prey items caught was recorded for successful dives. We then used various predicting variables derived from the other logging devices or from the environment to train a machine-learning algorithm to predict the outcome of each dive. Our results show that despite some limitations, the data collected from the footage was reliable as there was a high agreement from both annotators. We also demonstrate that it was possible to accurately predict the outcome of each dive from basic dive patterns and horizontal movement characteristics that have not been used for penguins previously. Finally, we discuss how video footage can help build more accurate habitat models and gain wider knowledge about predator behavior or prey distribution.

Prediction of inter-chain distance maps of protein complexes with 2D attention-based deep neural networks

Benchmark data sets of CDPred as described in Prediction of inter-chain distance maps of protein complexes with 2D attention-based deep neural networks Zhiye Guo1, Jian Liu1, Jeffrey Skolnick2, Jianlin Cheng1* 1 Department of Electrical Engineering and Computer Science, University of Missouri, Columbia, MO 65211 2 School of Biological Sciences, Georgia Institute of Technology, Atlanta, GA 30332-2000 *Corresponding author (chengji@missouri.edu) There is four test dataset in this package, each test dataset contains four different folders and one list file. The afpred_pdb includes all the corresponding monomer structures predicted by alphafold. The cdpred_output includes the prediction results of our tool CDPred for each dataset. The pre_gen_a3m includes the multiple sequence alignments file used by CDPred to generate prediction results. And the true_pdb includes the fasta file for the test dataset and its heavy atom distance map (h_dist) and carbon alpha distance map (real_dist) that extract from the native structure. HomoTest1: The homodimer test dataset contains 28 targets collect from CASP_CAPRI 10-13 HomoTest2: The homodimer test dataset contains 23 targets collect from CASP_CAPRI 13-14 HeteroTest1: The heterodimer test dataset contains 9 targets collect from CASP_CAPRI13-14 HeteroTest2: The heterodimer test dataset contains 55 targets collect from PDB bank 09-2021 to 11-2021

opencc-by-4.0Dec 2019View details →

A benchmark study of ab initio gene prediction methods in diverse eukaryotic organisms

G3PO (Gene and Protein Prediction PrOgrams) Benchmark was designed to represent many of the typical challenges faced by current genome annotation projects. The benchmark is based on a carefully validated and curated set of real eukaryotic genes from 147 phylogenetically disperse organisms (from human to protists).

Predicting Software Refactoring - joblib models

Zip file contains Machine Learning models trained as a partial result of our work described in our paper

Data for "Quantum-corrected thickness-dependent thermal conductivity in amorphous silicon predicted by machine learning molecular dynamics simulations"

This is the data set for the preprint <a href="https://arxiv.org/abs/2206.07605">arXiv:2206.07605</a> [cond-mat.mtrl-sci], obtained by the GPUMD code. Here are 6 directories.     1). NEMD     2). NEPpotential     3). PDOS     4). kappa-quenchRate     5). kappa-size     6). kappa-temperature      1). NEMD directory contains calculations of ballistic conductance using NEMD method, where 6 independent cycles are run to average. 2). NEPpotential directory is the trained NEP potential. 3). PDOS directory contains phonon density of states of a-Si samples generated by the quench rate of 10^{11} K/s. 4). kappa-quenchRate directory contains HNEMD calculations of a-Si samples which are prepared using melt-quench temperature protocols with the quench rates covering from 10^{11} to 5x10^{12} K/s. In each case, 3 independent cycles are run. 5). kappa-size directory contains HNEMD calculations based on different supercells. 6 independent cycles are run. 6). kappa-temperature directory contains HNEMD calculations of a-Si samples which are prepared for different targeted temperatures using slow quench rate of 10^{11} K/s.

Dataset for UAV velocity prediction_pt1

This file contains a dataset for UAV velocity prediction only in the Downtown West (Small City Downtown) map.

Predicting compound-protein interaction using hierarchical graph convolutional networks

This repository contains the datasets which are used in the article "Predicting Compound-Protein Interaction using Hierarchical Graph Convolutional Networks".

Towards Developing and Analysing The Metric-Based Software Defect Severity Prediction Model

This is a metric based approach to solve software defect severity prediction problem. In addition to that, this work proposes a new evaluation scheme that comprised of five metrics to analyze the performances.

Prediction Result of Next Monochrome Ionogram frame using LSTM-CNN Model

Video of next frame ionogram prediction result using simple LSTM-CNN model Paper published in The 9Th International Seminar on Aerospace Science and Technology ISAST 2022

Cell Graph data for predicting MSI vs. MSS from histological images

Abstract: The cell graph data extracted from histological images for predicting microstatellite status. This is a patch-level (patch refers to the sub-image tessellated from the whole-slide image) binary classification task: the two datasets classify images patches to either MSS (microsatellite stable) or MSIMUT (microsatellite instable or highly mutated). The original histology patches are available from [1].  Cell Graph Contraction: In this dataset, we establish a cell graph for each patch. The cell graph can represent the cell-cell interaction and the collection of cell graphs for all patches provide a precise characterization of tumor microenvironment. With only the availability of raw image patches, we leverage the nuclei regions segmented by a well-tuned CA2.5-Net [2] to extract the node features of each single nuclei node. And then we extract a total number of 94 pre-defined pathomics features for each nuclei region as the corresponding graph node feature. As the morphological signals are believed relative to cell-cell interplay, the cell-specific features which include the nuclei coordination, optical, and representations, then characterize the cell-level morphological behavior. We then calculate the pair-wise Euclidean distance between nuclei centroids to establish edges of a cell graph Reference For more details about the construction of cell-graph, please visit our paper: https://arxiv.org/abs/2206.07599. The associated codes and usage for the datasets are available at: https://github.com/yiqings/HEGnnEnhanceCnn. [1] Histological images for MSI vs. MSS classification in gastrointestinal cancer, FFPE samples. https://doi.org/10.5281/zenodo.2530835 [2] Huang, J., Shen, Y., Shen, D., & Ke, J. (2021, September). CA 2.5-Net Nuclei Segmentation Framework with a Microscopy Cell Benchmark Collection. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 445-454). Springer, Cham.

Towards understanding the importance of time-series features in automated algorithm performance prediction

merged_feature_importance.csv - CSV with feature importance values with different meta-models, forecasting algorithms, and feature importance methods computed on 30 different train/test splits. Catch22.csv - Catch22 features (raw time-series) Catch22Log.csv - Catch22 features (log time-series) Catch22Diff.csv - Catch22 features (differenced time-series) TSFresh.csv - TSFresh features (raw time-series) TSFreshLog.csv - TSFresh features (log time-series) TSFreshDiff.csv - TSFresh features (differenced time-series) mape.csv - sMAPE performance for all forecasting algoirthms

Direct prediction for carbapenemase-producing and colistin-resistant Klebsiella pneumoniae isolates from routine MALDI-TOF mass spectrum using machine learning

The emergence of carbapenem-nonsusceptible K. pneumoniae (CnSKP) leads a serious threat to patient survival and colistin resistance makes the treatment of CnSKP more difficultly. To make treatment strategy properly and quickly, we aimed to develop a rapid prediction method for CnSKP and colistin-resistant K. pneumoniae (ColRKP) based on the spectra of routine matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry (MALDI–TOF MS). The machine learning (ML) model for differentiating CnSKP and carbapenem-susceptible K. pneumoniae (CSKP) showed accuracy of 0.8869 and AUC of 0.9551; the model for ColRKP and colistin-intermediate K. pneumoniae (ColIKP) showed accuracy of 0.8361 and the AUC of 0.8447.