Predicting the biophysical and functional properties of proteins is essential for in silico protein design. Machine learning has emerged as a promising technique for such prediction tasks. However, the relative scarcity of in vitro annotations means that these models often have little, or no, specific data on the desired fitness prediction task. As a result of limited data, protein language models (PLMs) are typically trained on general protein sequence modeling tasks, and then fine-tuned, or applied zero-shot, to protein fitness prediction. When no task data is available, the models make strong assumptions about the correlation between the protein sequence likelihood and fitness scores. In contrast, we propose meta-learning over a distribution of standard fitness prediction tasks, and demonstrate positive transfer to unseen fitness prediction tasks. Our method, called Metalic (Meta-Learning In-Context), uses in-context learning and fine-tuning, when data is available, to adapt to new tasks. Crucially, fine-tuning enables considerable generalization, even though it is not accounted for during meta-training. Our fine-tuned models achieve strong results with 18 times fewer parameters than state-of-the-art models. Moreover, our method sets a new state-of-the-art in low-data settings on ProteinGym, an established fitness-prediction benchmark. Due to data scarcity, we believe meta-learning will play a pivotal role in advancing protein engineering.

Biological sequences encode fundamental instructions for the building blocks of life, in the form of DNA, RNA, and proteins. Modeling these sequences is key to understand disease mechanisms and is an active research area in computational biology. Recently, Large Language Models have shown great promise in solving certain biological tasks but current approaches are limited to a single sequence modality (DNA, RNA, or protein). Key problems in genomics intrinsically involve multiple modalities, but it remains unclear how to adapt general-purpose sequence models to those cases. In this work we propose a multi-modal model that connects DNA, RNA, and proteins by leveraging information from different pre-trained modality-specific encoders. We demonstrate its capabilities by applying it to the largely unsolved problem of predicting how multiple \rna transcript isoforms originate from the same gene (i.e. same DNA sequence) and map to different transcription expression levels across various human tissues. We show that our model, dubbed IsoFormer, is able to accurately predict differential transcript expression, outperforming existing methods and leveraging the use of multiple modalities. Our framework also achieves efficient transfer knowledge from the encoders pre-training as well as in between modalities. We open-source our model, paving the way for new multi-modal gene expression approaches.

Upcoming large satellite constellations and the advent of tighter steerable beams will offer unprecedented flexibility. Consequently, this will require resource management strategies to be operated in high-dimensional and dynamic environments, as existing satellite operators are unaccustomed to operational flexibility and automation. Frequency assignment policies have the potential to drive constellations’ performance in this new context, but are no exception to scalability and fast operation requirements. Most of existing frequency assignment methods fail to fulfill these requirements, or are unable to meet them without falling short on efficiency. In this paper we propose a new frequency assignment method that prioritizes operational requirements. We present an algorithm based on Integer Linear Programming that fully defines a frequency plan while respecting key system constraints such as handovers, interference, and gateway dimensioning. We can encode goals such as bandwidth maximization or power reduction and optimize plans according to such objectives. In our experiments on systems with 20 to 5,000 beams, we find this method allocates at least 100% more bandwidth and reduces power consumption by 40% compared to previous benchmarks. To ensure scalability, we also introduce an iterative approach of the formulation, which achieves substantial gains in runtime with minimal loss of performance.