Marie Lopez

loci. Finally, we identified a major contribution of western central African Bantu speakers to the ancestry of African Americans, whose genomes present no strong signals of natural selection. Together, these results highlight the contribution of Bantu-speaking peoples to the complex genetic history of Africans and African Americans.

Closing the gap between measurable genetic information and observable traits is a longstanding challenge in genomics. Yet, the prediction of molecular phenotypes from DNA sequences alone remains limited and inaccurate, often driven by the scarcity of annotated data and the inability to transfer learnings between prediction tasks. Here, we present an extensive study of foundation models pre-trained on DNA sequences, named the Nucleotide Transformer, ranging from 50M up to 2.5B parameters and integrating information from 3,202 diverse human genomes, as well as 850 genomes selected across diverse phyla, including both model and non-model organisms. These transformer models yield transferable, context-specific representations of nucleotide sequences, which allow for accurate molecular phenotype prediction even in low-data settings. We show that the developed models can be fine-tuned at low cost and despite low available data regime to solve a variety of genomics applications. Despite no supervision, the transformer models learned to focus attention on key genomic elements, including those that regulate gene expression, such as enhancers. Lastly, we demonstrate that utilizing model representations can improve the prioritization of functional genetic variants. The training and application of foundational models in genomics explored in this study provide a widely applicable stepping stone to bridge the gap of accurate molecular phenotype prediction from DNA sequence. Code and weights available at: https://github.com/instadeepai/nucleotide-transformer in Jax and https://huggingface.co/InstaDeepAI in Pytorch. Example notebooks to apply these models to any downstream task are available on https://huggingface.co/docs/transformers/notebooks#pytorch-bio.

Carbonic anhydrase (CA) enzymes, specifically membrane-bound isozymes CA IX and CA XII, underpin a pH-regulating system that enables hypoxic tumor cell survival and proliferation. CA IX and XII are implicated as potential targets for the development of new hypoxic cancer therapies. To date, only a few small molecules have been characterized in CA-relevant cell and animal model systems. In this paper, we describe the development of a new class of carbohydrate-based small molecule CA inhibitors, many of which inhibit CA IX and XII within a narrow range of low nanomolar K(i) values (5.3-11.2 nM). We evaluate for the first time carbohydrate-based CA inhibitors in cell-based models that emulate the protective role of CA IX in an acidic tumor microenvironment. Our findings identified two inhibitors (compounds 5 and 17) that block CA IX-induced survival and have potential for development as in vivo cancer cell selective inhibitors.

We report a study of xyloglucan (XG)-cellulose interactions made possible by the preparation of various well-defined cellulosic and xyloglucosidic substrates. Bacterial microcrystalline cellulose (BMCC) as well as cellulose whiskers (CellWhisk) were used as cellulosic substrates. Xyloglucosidic substrates were obtained from Rubus cells and Tamarindus indica seeds. Different primary structure characteristics of XGs such as the backbone length and the nature of the side chains, as well as their repartition, were considered in order to examine the influence of the primary structure on their interaction capacity. Two complementary approaches were carried out: first, the determination of adsorption isotherms and its associated models, and second, an enthalpic study using isothermal titration calorimetry (ITC). This study highlighted that an increase of XG interaction capacity occurred with increasing XG molecular weight. Furthermore, we determined that a minimum of 12 glucosyl residues on the backbone is required to observe significant interactions. Moreover, both the presence of trisaccharidic side chains with fucosyl residues and an increase of unsubstituted glucosyl residues enhanced XG-cellulose interactions. The evolution of adsorption isotherms with temperature and ITC measurements showed that two different processes were occurring, one exothermic and one endothermic, respectively. Although the presence of an exothermic interaction mechanism has long been established, the presence of an endothermic interaction mechanism has never been reported.