"Over the last years we collected experimental data from hundreds of individuals
using different experimental techniques - how can we combine and analyze these data to obtain novel insights?"
Addressing research questions in the life sciences today often requires integrating and analyzing
vast amounts of data. Transforming high-throughput experimental data into meaningful representations,
detecting patterns in high-dimensional noisy data, finding complex correlations between biological
entities and connecting them to phenotypes: transforming data into knowledge requires a solid understanding
of data characteristics and semantics. Genevention draws on long-standing experience in deep data analysis
and provides computational analysis solutions for a broad range of life science problems. We develop
cutting-edge big data processing and analysis solutions – highly integrated or completely portable
using state-of-the-art container technologies such as Docker. Benefit from our data mining, machine
learning and artificial intelligence expertise reveal the
knowledge in your data!
Click Image to Test Biomarker Showcase Platform
Cutting-edge, automated, reproducible and portable bioinformatics workflows are needed that rely
on scientific best practices and state-of-the-art software development methods.
Our biomarker identification toolbox (BIT) performs large-scale (differential) expression,
infection and mutation analyses from the complete range of publicly available (small)RNA datasets
that have been semantically annotated in-house by biomedical experts.
In the adjacent applet you can explore a small RNA signature identified by BIT for a certain cancer
type by overlapping differentially expressed genes from annotated expression datasets.
BIT is fast, accurate, flexible and easy to use - contact us
for how BIT can help to solve your bioinformatics challenges.
Deep Data Analytics
Generating realistic biomedical data using Deep Learning techniques:
conditional Generative Adversarial Networks (cGANs) are used to generate in silico single-cell
expression profiles that cannot be distinguished from real profiles. GANs learn complex gene-gene
dependencies from multi cell type complex samples and use this information to generate realistic
cells of defined type. The technology can be used for any data domain where a low amount of
observations is available, but more samples are desirable – to save animal lives and money and
to increase reproducibility of results. (Paper)
Evaluation of scGANs in PBMC cells -
distributions of real and simulated single cell transcriptome profiles
Realistic generation of single cell
transcriptome profiles for cell types is possible from very few examples