DUNE: Deep feature extraction by UNet-based Neuroimaging-oriented autoEncoder

A versatile neuroimaging encoder that captures brain complexity across multiple diseases: cancer, dementia and schizophrenia.

Overview

DUNE (Deep feature extraction by UNet-based Neuroimaging-oriented autoEncoder) is a neuroimaging-oriented deep learning model designed to extract deep features from multisequence brain MRIs, enabling their processing by basic machine learning algorithms. This project provides an ...

Description

The Settlement Delineation and Analysis (SDA) workflows generates a settlement network from geospatial settlement data. It can process geotiff and shapefile inputs and was originally designed to operate on the World Settlement Footprint dataset. Through multiple workflow stages, a settlement network is constructed, contracted (i.e. clustered) and ultimately analysed with centrality measures. The output shapefile stores the ...

Colorectal-cancer-detection-using-ColoPola-dataset

Methods

We trained and tested three models from scratch (CNN, CNN_2 and EfficientFormerV2) and two pretrained models (DenseNet121 and EfficientNetV2-m) to classify the colorectal cancer using the ColoPola dataset.

ColoPola dataset

The dataset consists of 572 slices (specimens) with 20,592 images. There are 284 cancer samples and 288 normal samples. This dataset can download from Zenodo repository. ...

CryoDataBot

CryoDataBot is an automated pipeline designed to streamline dataset generation for cryogenic electron microscopy (cryoEM)-based atomic model building. It supports large-scale AI training and benchmarking by providing standardized tools for data retrieval, preprocessing, labeling, and quality control. CryoDataBot enables flexible configurations for diverse biomolecular structures, improves modeling reproducibility, and facilitates retraining of AI models such as U-Net and CryoREAD. ...

[![Cite with ...

...

This workflow extracts protein-coding sequences from whole genome sequencing (WGS) data obtained from the European Nucleotide Archive (ENA). It automates the preprocessing, annotation, and selection of relevant protein sequences using tools such as Prokka, FASTA-to-Tabular, and pattern-based selection. The resulting dataset supports downstream analyses including comparative genomics, phylogenetics, and functional annotation.

GenErode pipeline

GitHub repository for GenErode, a Snakemake workflow for the analysis of whole-genome sequencing data from historical and modern samples to study patterns of genome erosion.

Documentation

The full pipeline documentation can be found on the repository wiki.

Citation

If you've ...

Automated image processing from movies to 3D reconstruction. Includes CTF estimation, quality filters, box size estimation, picking model training, three 2D classification jobs (25k, 50k, and 100k particles), and 3D processing of the first 200k particles. Multiple 3D models are generated without symmetry, with automatic 2D and 3D class selection, using Relion.

Automated image processing from movies to 3D reconstruction. Includes CTF estimation, quality filters, box size estimation, picking model training, three 2D classification jobs (25k, 50k, and 100k particles), and 3D processing of the first 200k particles. Multiple 3D models are generated without symmetry, with automatic 2D and 3D class selection, using Relion.