The performance differential of large language models (LLM) between languages hinders their effective deployment in many regions, inhibiting the potential economic and societal value of generative AI tools in many communities. However, the development of functional LLMs in many languages (i.e., multilingual LLMs) is bottlenecked by the lack of high-quality evaluation resources in languages other than English. Moreover, current practices in multilingual benchmark construction often translate English resources, ignoring the regional and cultural knowledge of the environments in which multilingual systems would be used. In this work, we construct an evaluation suite of 197,243 QA pairs from local exam sources to measure the capabilities of multilingual LLMs in a variety of regional contexts. Our novel resource, INCLUDE, is a comprehensive knowledge- and reasoning-centric benchmark across 44 written languages that evaluates multilingual LLMs for performance in the actual language environments where they would be deployed.

MOTIVATION: Protein phosphorylation is a ubiquitous regulatory mechanism that plays a central role in cellular signaling. According to recent estimates, up to 70% of human proteins can be phosphorylated. Therefore, the characterization of phosphorylation dynamics is critical for understanding a broad range of biological and biochemical processes. Technologies based on mass spectrometry are rapidly advancing to meet the needs for high-throughput screening of phosphorylation. These technologies enable untargeted quantification of thousands of phosphorylation sites in a given sample. Many labs are already utilizing these technologies to comprehensively characterize signaling landscapes by examining perturbations with drugs and knockdown approaches, or by assessing diverse phenotypes in cancers, neuro-degerenational diseases, infectious diseases and normal development. RESULTS: We comprehensively investigate the concept of 'co-phosphorylation' (Co-P), defined as the correlated phosphorylation of a pair of phosphosites across various biological states. We integrate nine publicly available phosphoproteomics datasets for various diseases (including breast cancer, ovarian cancer and Alzheimer's disease) and utilize functional data related to sequence, evolutionary histories, kinase annotations and pathway annotations to investigate the functional relevance of Co-P. Our results across a broad range of studies consistently show that functionally associated sites tend to exhibit significant positive or negative Co-P. Specifically, we show that Co-P can be used to predict with high precision the sites that are on the same pathway or that are targeted by the same kinase. Overall, these results establish Co-P as a useful resource for analyzing phosphoproteins in a network context, which can help extend our knowledge on cellular signaling and its dysregulation. AVAILABILITY AND IMPLEMENTATION: github.com/msayati/Cophosphorylation. This research used the publicly available datasets published by other researchers as cited in the manuscript. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Genome-wide association studies show that variants in individual genomic loci alone are not sufficient to explain the heritability of complex, quantitative phenotypes. Many computational methods have been developed to address this issue by considering subsets of loci that can collectively predict the phenotype. This problem can be considered a challenging instance of feature selection in which the number of dimensions (loci that are screened) is much larger than the number of samples. While currently available methods can achieve decent phenotype prediction performance, they either do not scale to large datasets or have parameters that require extensive tuning. RESULTS: We propose a fast and simple algorithm, Macarons, to select a small, complementary subset of variants by avoiding redundant pairs that are likely to be in linkage disequilibrium. Our method features two interpretable parameters that control the time/performance trade-off without requiring parameter tuning. In our computational experiments, we show that Macarons consistently achieves similar or better prediction performance than state-of-the-art selection methods while having a simpler premise and being at least two orders of magnitude faster. Overall, Macarons can seamlessly scale to the human genome with ∼107 variants in a matter of minutes while taking the dependencies between the variants into account. AVAILABILITYAND IMPLEMENTATION: Macarons is available in Matlab and Python at https://github.com/serhan-yilmaz/macarons. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Phenotypic heritability of complex traits and diseases is seldom explained by individual genetic variants identified in genome-wide association studies (GWAS). Many methods have been developed to select a subset of variant loci, which are associated with or predictive of the phenotype. Selecting connected SNPs on SNP-SNP networks have been proven successful in finding biologically interpretable and predictive SNPs. However, we argue that the connectedness constraint favors selecting redundant features that affect similar biological processes and therefore does not necessarily yield better predictive performance. In this paper, we propose a novel method called SPADIS that favors the selection of remotely located SNPs in order to account for their complementary effects in explaining a phenotype. SPADIS selects a diverse set of loci on a SNP-SNP network. This is achieved by maximizing a submodular set function with a greedy algorithm that ensures a constant factor approximation to the optimal solution. We compare SPADIS to the state-of-the-art method SConES, on a dataset of Arabidopsis Thaliana with continuous flowering time phenotypes. SPADIS has better average phenotype prediction performance in 15 out of 17 phenotypes when the same number of SNPs are selected and provides consistent improvements across multiple networks and settings on average. Moreover, it identifies more candidate genes and runs faster.