I forskargruppen använder vi beräkningsverktyg för att ställa stora frågor gränsöverskridande frågor om till exempel identitet, förhållandet mellan genoptyp och fenotyp och evolution.

Vi drivs av fascinationen och intresset för att använda beräkningsverktyg för att ställa stora, multidisciplinära frågor om identitet, evolution och samband mellan genotyp och fenotyp. Vår forskning är inte begränsad till specifika organismgrupper.

Eran Elhaik lab

Our lab group has broad expertise that allows us to tackle a wide range of questions within and outside of medical and life sciences in collaborations across the globe. Our main projects revolve around population, medical, and evolutionary genomics, as well as metagenomics. Our current study systems include ancient and modern humans, plants, and the microbiome. Our lab also leads the development and integration of Artificial Intelligent (AI) tools in omic studies.

Nikolay Oskolkov, PhD, senior bioinformatician, NBIS SciLifeLab

My research interests focus around computational biology, mathematical statistics, machine learning and data science. I am involved in different projects varying from biomedical analysis and single-cell omics to evolutionary biology and ancient DNA.

Postdocs

Sara Behnamian

Here we list some of the major research current projects in the research group. Please contact us for more information and opportunities.

The Microbiome in our surrounding environment – ecology, forensics, paleogenomics, and legal implications

In the past few years, studies have characterized the microbiota and metagenome of urban environments and transit systems and demonstrated that some species are specific to certain areas of a city. These “molecular echoes” of the environment allow the development of biogeographical applications, which allow forensic capabilities and have legal implications. These studies add to our accumulated knowledge of the microbiome in the soil and marine environment, as we all as the microbiome on and within our bodies and help to understand the contact points between humans and the microbial world. As part of the Metagenomics and Metadesign of the Subways and Urban Biomes (MetaSUB) International Consortium of over 100 cities worldwide, we collaborate with experts across many fields, including genomics, data analysis, engineering, public health, and architecture. The ultimate goal of the MetaSUB Consortium is to improve city utilization and planning through the detection, measurement, and design of metagenomics within urban environments. Our lab is developing Artificial Intelligence (AI) tools to trace microbiome samples geographically with great accuracy and understand the spread of antimicrobial resistance genes (AMR). We are also studying the ancient and modern microbiome in archeological sites and historical monuments and what they can tell us about ancient and modern societies.
Contact: Eran Elhaik

Developing Artificial Intelligence (AI) applications for ancient reconstructions

Reconstructions of past events, whether for people, animals, plants, genes, or alleles using ancient DNA (aDNA), require answering two questions: When and Where. When is the piece of DNA from, and Where is it from? Without answering these two questions, no paleogenomic reconstructions can take place; we will resort to coalescent methods and their fallacies. Where is seemingly simple, we usually know where the bone was found, but we never know where it migrated from. Answering when the sample is from is also seems obvious. However, a closer look reveals that only half of the aDNA data are radiocarbon dated and, despite decades of development, radiocarbon dating remains imperfect. Therefore, current reconstructions primarily rely on inaccurate methods borrowed from modern-day DNA tuned to fit some known historical scenarios. Our lab set to change that by developing Artificial Intelligence (AI) tools to date ancient genomes and predict their historical migration routes – solely from DNA data. Using our tools, scientists can reconstruct the history of populations, single individuals, animals, microorganisms, genes, haplogroups, and even alleles. Our tools allow a vision of the past at a fine-tuned level.
Contact: Eran Elhaik

Modeling population structure of modern-day populations

Knowledge about ancestry and origins is central to the study of populations. Our present-day genetic diversity was shaped by biological and demographic events that marked their signatures in the genome. My lab asks questions like: how do we predict the geographical origin of samples based on their DNA? How do we reconstruct the migration route over time? How do we use this information to address population stratification and improve the accuracy of disease and clinical studies? 
We previously showed that geographical origin could be accurately inferred solely from genomic data (our GPS tool). We also developed a tool that capitalizes on ancestry information to improve the accuracy of case-control matches (PaM). We also showed that our technology could address unresolved questions concerning the ancient origins of populations and, in one case, uncovered ancient Ashkenaz, where the genome of Ashkenazic Jews has originated. Many of our tools were commercialized and are available to the general public. We continue to improve the accuracy of our tools and study the origins of populations of interest.
Contact: Eran Elhaik

Identifying genes involved in complex disorders (mental disorders, multiple myeloma, ALS, and Spina Bifida)

Multifactorial disorders exhibit a complex architecture consisting of disease-causing variants and environmental modifying factors. Classic complex disorder studies focus on families with more modern approaches relying on Big Data from large populations, which subjects them to the risks of population stratification. Even after addressing these risks, it remains challenging to develop predictive models for complex diseases whose etiology may be poorly understood. For that, our lab develops Artificial Intelligence (AI) based tools that integrate biomedical Big Omics Data with population genetic models to predict the diagnosis and risk for the disease. 
Contact: Eran Elhaik