Microscopy, the smart way: Christian Conrad receives BIH Professorship for Intelligent Imaging

Researchers today often do not look through the microscope themselves: With the vast number of cells, tissue sections and tumor samples, there is simply too much for the human eye to spot and record. It was for this reason that Christian Conrad introduced machine learning into the field of medical imaging early on. With the help of artificial intelligence (AI), thousands of samples and cells can now be automatically analyzed using high-throughput methods. In addition, the combination of microscopy, AI and new single-cell sequencing methods opens up entirely new possibilities for discovering the function of cells in their tissue context. At the beginning of the year, Christian Conrad assumed the BIH Professorship for Intelligent Imaging at the Digital Health Center of the Berlin Institute of Health (BIH) and Charité – Universitätsmedizin.

“Personalized medicine relies on an accurate diagnosis,” explains Professor Christopher Baum, of the BIH Board of Directors and Chief Translational Research Officer of Charité. “To ensure this, it is crucial that we are able to decipher the molecular events that take place in human tissue. Using state-of-the-art sequencing technologies, it is possible to analyze tissue samples three-dimensionally at the single-cell level, and thus obtain an unprecedented wealth of information on the course of disease. We at the BIH are therefore delighted that, with Christian Conrad, we have gained a recognized expert in this important field.”

Conrad’s interested in microscopy began early on in his career: “Even while I was writing my doctoral thesis at the German Cancer Research Center in Heidelberg, I was fascinated by the ability to render cell function visible with the help of different fluorescent stains.” In recent years, fluorescence has provided more and more ways to obtain information about proteins or cell gene patterns. “With regard to cancer cells, this is extremely important information that can also have therapeutic use.”

Automated tumor analysis

At the Digital Health Center of the BIH and Charité, scientists led by Christian Conrad are also working with cancer cells – now with the aim of finding out which course of treatment might be the best for the individual patient. To do this, they obtain primary tissue samples from tumors that have been removed. “These are often only very small amounts – not enough for us to test the effectiveness of 50 different drugs,” explains Conrad. “We therefore use the stem cells contained in this tissue to cultivate small, three-dimensional tissue pieces known as organoids, and use these to carry out the tests. Under the microscope, we can observe whether a substance has caused the tumor to shrink, whether it kills tumor cells or whether it has no effect.”

But because there are so many samples to be tested, clinically and diagnostically, the researchers are working to refine the automated microscopy procedure: “We use culture plates that can hold up to 100 organoids at a time. These are automatically analyzed under the microscope and the data is promptly evaluated directly in the computer. Our goal is to use artificial intelligence to develop models that can provide information about which drug is likely to be most effective against which tumor,” explains Conrad.

Analyzing genes in the tissue itself

In parallel, the scientists are also analyzing the genetic expression of these mini-tumors. This provides them with insight into the genetic changes contained in the tumor and allows them to predict which targeted drugs are most likely to be active in combating the tumor cells. “In most cases, our prediction based on the sequencing data corresponds to the observations made under the microscope, but sometimes it does not,” reports Conrad. “We want to understand the relationship between morphology – i.e., the form of the organoids – and gene expression.”

Special single-cell methods now make it possible to decode the messenger RNA (mRNA) in each individual cell. These gene transcripts reveal, in detail, which genes are active in which cell. Conrad and his team are developing methods that would allow this single-cell sequencing method to be used directly under the microscope (in situ). “It is precisely because tumors are usually a highly heterogeneous mixture of tumor cells, connective tissue cells, immune cells and blood vessels that it is advantageous to know which genes are active in which exact cell and the cell types that are involved. This way, we can determine whether immune cells are actively fighting the tumor, how fast the tumor cells are multiplying, or whether blood vessels could grow into the tumor.”

Naturally, these in situ single-cell analyses are not only interesting for cancer research: “We have also been approached by cardiologists who want to analyze tissue from cardiac catheterizations. And of course, given the current situation, pulmonologists and virologists are also interested in finding out which cells the Sars-CoV-2 viruses thrive in.” The method of spatially localizing RNA, known as resolved spatial transcriptomics, was just named “Method of the Year 2020” by the journal Nature Methods. Conrad is therefore working in a field with great promise for the future.

Christian Conrad was born in Hamburg in 1968, studied biology in Freiburg, and then worked for four years in a start-up company developing automated microscopy systems in cytogenetics and pathology. For his doctorate, which he obtained at the German Cancer Research Center in Heidelberg under Professor Roland Eils, he returned to academia to study automated microscopy and pattern recognition. He then went on to head various research groups on automated microscopy in his field of expertise, intelligent imaging, first at the European Molecular Biology Laboratory (EMBL) and later at BioQuant – Center for Quantitative Analysis of Molecular and Cellular Biosystems at Heidelberg University. In 2018, he and Professor Eils joined the Digital Health Center of the BIH and Charité in Berlin. Conrad is married and has three daughters.