Anduril (ANalysis of Data Using Rapid Integration of aLgorithms) is an opensource component-based workflow framework for systematic, flexible and efficient scientific data analysis, particularly in the field of high-throughput experiments in biomedical research. It includes components for expression, SNP, ChIP-seq and exon microarray analysis.
Anima is an image analysis workflow platfrom built on top of Anduril. Anima acts as a superplatform for the existing imaging applications, and can run components and software written many languages. In addition Anima provides API to the popular batch imaging environments: Fiji, CellProfiler and ImageMagick.
GROK (Genomic Region Operation Toolkit) is "Swiss Army knife" library for processing genomic interval data. GROK operates ongenomic regions, annotated chromosomal intervals that represent sequencing short reads, gene locations, ChIP-seq peaks or other genomic features. Applications of GROK include file format conversions, set operations, overlap queries, and filtering and transformation operations.
POIBM (Poisson batch correction trough sample matching) was developed for batch correction of RNA-seq data. The method finds both batch factors and matching samples between two sets of heterogeneous RNA-seq data, which can be used for correction. The data sets do not need to be prelabeled and no overlapping samples are required as the method relies on finding matching phenotypes among the various subgroups in the data.
DMML (Differential Methylation Maximum Likelihood) is method based on a latent stochastic model which allows comparing DNA methylomes at a single-nucleotide resolution between two or more tumor samples with different, unknown tumor purities. The method performs accurately without a normal cell control sample or prior tumor purity estimates, and accounts for spatially co-methylated cytosines, improving accuracy at lower coverage sites.
QuantISH is a new open-source software that offers a comprehensive method for quantifying signals in individual cells from images of singleplex, chromogenic RNAscope. It has a modular design that makes it adaptable to various image and sample types and staining protocols. The software's modularity allows for straightforward upgrades to more advanced methods when they become available. QuantISH goes beyond simply estimating average expression; it can also calculate the variability of expressions. In addition to being the first software of its kind, QuantISH is also easy to use and can be applied to a wide range of image analyses with only minor modifications.
PRISM is a latent statistical framework that resolves the issue of sample heterogeneity and systematic changes in the composition of bulk RNA samples. Using PRISM, it’s possible to simultaneously extract the sample composition and cell-type-specific whole-transcriptome profiles tailored to each sample. The validity of the tool was verified in independent ovarian cancer and melanoma cohorts through whole-genome sequencing and RNA in situ hybridization experiments. Ultimately, PRISM allows for improved prediction of patient response to treatment compared to traditional approaches using composite raw bulk data.
HEIP is a Hematoxylin and eosin (H&E) Image Processing pipeline designed to extract cell-level information from H&E whole slides. It incorporates various steps and related tools, including HE-image patching, pre-processing, cell segmentation, classification, and downstream feature extraction. The pipeline is entirely written in Python, offering modularity and ease of adding or modifying existing steps.
GenomeSpy presents a grammar-based, GPU-accelerated visualization toolkit designed for genomic data. Much like ggplot2 in bioinformatics, GenomeSpy empowers users to swiftly construct customized visual paradigms in cases where standard solutions prove insufficient. Its utility spans diverse applications, including creating custom genome browsers for exploring multidimensional datasets. Additionally, GenomeSpy facilitates seamless investigation of sizable sample repositories, offering robust sorting, filtering, grouping, and aggregation capabilities.