Using our automated high-throughput zebrafish embryo based service platform, you explore advantages of the time and cost-efficient alternatives to cell and mammal-based testing.
CRISPR/Cas9 mutagenesis and transgenesis
CRISPR/Cas9 knock-out fish are created by introducing small insertions or deletions early in the coding sequence of the gene leading to the occurrence of the premature STOP codon, or by deletion of the gene promotor region. Similarly, CRISPR/Cas9 can be used for generation of the specific transgenic lines.
Out team has experience of generating and evaluating of over 150 gene knock-outs and several transgenic lines for gene function and disease modelling purpose.
Fast development, small size and optical transparency at early developmental stages make zebrafish model widely used for high-throughput chemical screenings in 96-well format. High range of the fluorescent and luminescence-based zebrafish reporter lines provide a strong base for the visualization and evaluation of the screens focused on the tissue specificity.
DanioReadout offers wide range of the phenotype-based readouts of the chemical screens based on the in-house developed image analysis methods.
Zebrafish is being used as a model to study cancer or patient derived cancer cell transplants in the expanding field of personalized medicine.
We are ready to participate in studies involving cancer cell lines and patient derived xenotransplantations.
Data Acquisition and Analysis
Ubiased phenotyping in zebrafish using data-driven AI
BIIF and GEZ (2020-present)
Artificial intelligence and deep learning has had a great impact on the field of image analysis. However, these methods need a large amount of data to be used for learning. Due to this requirement the zebrafish community has not yet seen the full potential of today’s AI. With access to our high-throughput systems here in Uppsala we have a unique position to combine zebrafish research with the current technology advances in AI. In this project we aim to use our large dataset of images of different phenotypes to develop novel AI methods for ubiased phenotypic screening. The aim here is not to define features to look for but rather without any a priori information detect features that differ between treatment groups using AI.
Morphometric measurements of the zebrafish using deep-learning
BIIF and GEZ (2019-present)
In this project we aim to have a segmentation tool for morphometric measurments of zebrafish images acquired in the VAST system. We have been generating a large amounts of data using our High-Througphut system during the last 2 years. We are now using these thousands of images of zebrafish larvae of ages between 3-10 days post fertilization (dfp) to train our neural network to robustly segment zebrafish in our images. The next step we are working on is to train the network on more morphological structures within the fish. This will provide researchers with more and better quantifation metrics of the zebrafish development.
Automatic high-throughput screening of the zebrafish brain using VAST
BIIF and GEZ (2019-present)
We have establish an automatic high-throughput phenotypic screening pipeline using the VAST system. The system can automatically load fish and acquire images in both bright-field and fluorescence. Using a convolutional neural network we segment the bright-field images to get an estimate of length and shape. For the fluorescence data we rotate the sample and acquire high-resolution images from different angles. For each view, images are aligned using an intensity-based registration algorithm and the results can be used to compare average patterns and get regions of statistically significant differences. In addition, utilizing multiple views for each fish enables us to identify the 3D position of the detected phenotypes based on the optical projection tomography technique. We have evaluated our pipeline using a CRISPR/Cas9 dbx1a;dbx1b double mutant line in the transgenic Tg(dat:EGFP) background and detected multiple statistically significant phenotypes in the zebrafish brain including pretectum and ventral diencephalon.
Optical projection tomography of zebrafish.
BIIF and GEZ (2018-present)
Optical projection tomography (OPT) is a 3D imaging alternative to conventional microscopy which allows imaging of millimeter-sized object with isotropic micrometer resolution. The size and optical transparency of the embryo and larva makes them well suited for imaging using OPT. We have developed an OPT plattform for rapid 3D imaging of zebrafaish larva. We have also developed computational methods for alignment of 3D data and methods for detection of statistically significant phenotypes.
- "zOPT: an open source optical projection tomography system and methods for rapid 3D zebrafish imaging,", Hanqing Zhang, Laura Waldmann, Remy Manuel, Henrik Boije, Tatjana Haitina, and Amin Allalou, Biomed. Opt. Express 11, 4290-4305 (2020)
- "The role of Gdf5 in the development of the zebrafish fin endoskeleton", Laura Waldmann, Jake Leyhr, Hanqing Zhang, Amin Allalou, Caroline Ohman-Magi, Tatjana Haitina, https://www.biorxiv.org/content/10.1101/2021.01.31.428959v1 (2021)
- "The Broad Role of Nkx3.2 in the Development of the Zebrafish Axial Skeleton", Laura Waldmann, Jake Leyhr, Hanqing Zhang, Caroline Öhman-Mägi, Amin Allalou, Tatjana Haitina https://www.biorxiv.org/content/10.1101/2020.12.30.424496v1 (2021)