Bernardo de Almeida wins 2024 Denise P. Barlow Award
Bernardo de Almeida, a recent PhD graduate from Alexander Stark's lab at the IMP, has been honoured with the Denise P. Barlow Award for his exceptional contributions to the computational modelling and design of enhancers – DNA sequences that are crucial for gene regulation. De Almeida’s doctoral thesis was recognised for its significant impact on our understanding of enhancers’ function, which further sheds light on the fundamental mechanisms of gene expression.
During his PhD with Alexander Stark at the IMP, Bernando de Almeida developed and spearheaded the application of deep and transfer learning – a form of machine learning in artificial intelligence – to the study of genetic enhancers. With his contributions he achieved a key goal that had been pursued in the field for decades: the computational modelling and design of enhancer sequences. His groundbreaking PhD thesis now earned him the 2024 Denise P. Barlow Award.
‘Enhancers’ are genomic DNA segments that play a critical role in switching on genes. They are responsible for activating specific genes at the right time and in the right cellular context, thereby influencing various biological processes such as development, differentiation, and response to environmental stimuli. During his time as a PhD student in the lab of Alexander Stark, de Almeida focused on addressing the question of how gene-regulatory information is stored within the DNA sequences of enhancers.
De Almeida realised that recent progress in experimental methods – combined with the technical breakthroughs in artificial intelligence (AI) – now allowed new approaches to this question. Therefore, he implemented the use of the most advanced deep learning models and applied them to enhancer-activity datasets previously generated in collaboration with Franziska Reiter, then also a PhD student in the lab of Alexander Stark.
He developed a deep-learning model called DeepSTARR, to predict enhancer activity in the cells of the fruit fly Drosophila melanogaster, a widely used model organism in developmental biology. De Almeida identified key elements within enhancer sequences, revealing important ‘code rules’ that characterise them. This allowed the design of synthetic enhancers that functioned at the different intended activity levels – weak, medium, or strong. His groundbreaking work was published in 2022 and earned him the Life Science Research Award Austria 2022.
Building onto this, De Almeida led the creation of a powerful system to predict enhancer activity in living Drosophila embryos using a deep and transfer learning model. This model was first trained on large sets of data from previous studies, specifically on genome-wide DNA sequences and corresponding DNA accessibility data – a predictor of enhancers. This first model was then used to fine-tune a second model, which learned to directly link DNA sequences to specific enhancer activities.
With this approach the model was able to predict enhancer activity for five types of tissues in fruit fly embryos: the central nervous system, epidermis, gut, and muscle. The researchers were then able to test 40 different synthetic enhancers that were designed computationally in living fruit fly embryos. The result was active enhancers that switched on gene expression in the targeted tissues.
De Almeida’s work constituted the first targeted design of synthetic enhancers for selected tissues, a true breakthrough achievement that was published in Nature at the end of 2023. In recognition of this pioneering discovery, De Almeida received the Denise P. Barlow Award yesterday, in a ceremony held in the Festival Hall of the Austrian Academy of Sciences.
“It’s an honour to receive an award named after Denise Barlow who pioneered visionary research in epigenetics at the IMP 30 years ago,” says De Almeida. “We are now at another turning point in the life sciences, using the tool of artificial intelligence to drive our knowledge of gene regulation forward."
About the Denise P. Barlow Award
Denise Barlow had a prolific scientific career, mostly remembered for her contributions to early epigenetics. A detailed account of the work she had done while starting her lab at the IMP can be found on this website. Barlow’s career, however, continued to span four institutions in Vienna: following the IMP, where she was a group leader, it also included the Max Perutz Labs, where she was a professor, IMBA, which provided laboratory space for some time, and CeMM, where she was a principal investigator until her retirement.
Following Barlow’s untimely passing in 2017, these four institutions launch an annual award in her name to merit exceptionally interesting PhD studies carried out in one of the four institutions. Specifically, the Denise P. Barlow Award intends to promote the academic career of young scientists, strategically addressing the transition of graduated PhD students to postdocs, by offering a small financial help that will alleviate the costs associated with relocation or career development.
The “Denise P. Barlow Award” amounts to 5,000 Euro. After two years of postdoc research, awardees are invited to give a talk, and while in Vienna, participate in the award of the next winner. The topic of the thesis can vary and cover basic cell biological, biochemical, molecular biological, structural, and computational work, with an emphasis on insight on new biological mechanisms, something Denise P. Barlow was fond of. The award selection committee comprises seven people, including the four scientific directors of the awarding institutes.
About Bernardo de Almeida
Bernardo de Almeida studied Biomedical Sciences and Oncobiology at the University of Algarve (Portugal) and worked on several computational biology projects, at his home university, in Lisbon (Portugal), and at the Sanger Institute in Cambridge (UK). De Almeida then joined the IMP as a PhD student in the lab of Alexander Stark, with a project focused on how gene-regulatory information is encoded in the DNA sequences of genomic regulatory elements. Bernardo de Almeida’s academic journey is adorned with the 2022 Life Science Research Award Austria, and the 2023 Vienna BioCenter PhD Award – both earned for his contributions to the development of the deep-learning model DeepSTARR for the prediction of enhancer sequences. De Almeida is currently holding a position as a Research Scientist in artificial intelligence and biology at the company InstaDeep in Paris, France.
Further reading
Lab of Alexander Stark
Artificial Intelligence cracks code of gene regulation
Harnessing artificial intelligence to predict and control gene regulation