NAR Genomics and Bioinformatics最新文献

In bacteria, high A and low G content of the 5' end of the coding sequence (CDS) promotes low RNA stability, facilitating ribosomal initiation and subsequently a high protein to transcript ratio. Additionally, 5' NGG codons are suppressive owing to peptidyl-tRNA drop off. It was, therefore, surprising that the first large-scale transgene experiment to interrogate the 5' effect by codon randomization found the NGG, G-rich codon AGG to be the most associated with high transgene output. Why is this? We show that this is not replicated in other large transgene datasets, where AGG and NGG are associated with low efficiency. More generally, there is limited agreement between the first experiment and others. This we find to be a consequence of non-random construct design. In constructs of the first experiment, AGG disproportionately occurs with non-AGG codons associated with low stability and high protein output, making AGG's association with high output an artefact. While translationally non-optimal codons like AGG are conjectured to slow ribosomes for orderly initiation, we find that in the less biased constructs high, not low, translational adaptation in the first 10 codons is (weakly) predictive of higher translational efficiency. These results have implications for both transgene and experimental design.

While monoclonal antibody-based targeted therapies have substantially improved progression-free survival in cancer patients, the variability in individual responses poses a significant challenge in patient care. Therefore, identifying cancer subtypes and their associated biomarkers is required for assigning effective treatment. In this study, we integrated genotype and pre-treatment tissue RNA-seq data and identified biomarkers causally associated with the overall survival (OS) of colorectal cancer (CRC) patients treated with either cetuximab or bevacizumab. We performed enrichment analysis for specific consensus molecular subtypes (CMS) of CRC and evaluated differential expression of identified genes using paired tumor and normal tissue from an external cohort. In addition, we replicated the causal effect of these genes on OS using a validation cohort and assessed their association with The Cancer Genome Atlas Program data as an external cohort. One of the replicated findings was WDR62, whose overexpression shortened OS of patients treated with cetuximab. Enrichment of its overexpression in CMS1 and low expression in CMS4 suggests that patients with the CMS4 subtype may derive greater benefit from cetuximab. In summary, this study highlights the importance of integrating different omics data for identifying promising biomarkers specific to a treatment or a cancer subtype.

Large-scale quantitative studies have identified significant genetic associations for various neurological disorders. Expression quantitative trait locus (eQTL) studies have shown the effect of single-nucleotide polymorphisms (SNPs) on the differential expression of genes in brain tissues. However, a large majority of the associations are contributed by SNPs in the noncoding regions that can have significant regulatory function but are often ignored. Besides, mutations that are in high linkage disequilibrium with actual regulatory SNPs will also show significant associations. Therefore, it is important to differentiate a regulatory noncoding SNP with a nonregulatory one. To resolve this, we developed a deep learning model named Neur-Ally, which was trained on epigenomic datasets from nervous tissue and cell line samples. The model predicts differential occurrence of regulatory features like chromatin accessibility, histone modifications, and transcription factor binding on genomic regions using DNA sequence as input. The model was used to predict the regulatory effect of neurological condition-specific noncoding SNPs using in silico mutagenesis. The effect of associated SNPs reported in genome-wide association studies of neurological condition, brain eQTLs, autism spectrum disorder, and reported probable regulatory SNPs in neurological conditions were predicted by Neur-Ally.

[This corrects the article DOI: 10.1093/nar/lqae156.].

Antibody production by B cells is essential for protective immunity. The clonal selection theory posits that each mature B cell has a unique immunoglobulin receptor generated through random gene recombination and, when stimulated to differentiate into an antibody-secreting cell, has the capacity to produce only a single antibody specificity. It follows from this 'one-cell-one-antibody' dogma that single-cell RNA-seq profiling of antibody-secreting cells should find that each cell expresses only a single form of each of the immunoglobulin heavy and light chains. However, when using GRCh38 as the genome reference, we found that many antibody-secreting cells appeared to express multiple immunoglobulin isotypes. When the newly published T2T-CHM13 genome was used instead as the genome reference, every antibody-secreting cell was found to express a unique isotype, and read mapping quality was also improved. We show that the superior performance of T2T-CHM13 was due to its European origin matching the genetic background of the query samples. On the other hand, T2T-CHM13 failed to appropriately fit the 'one-cell-one-antibody' dogma when applied to data derived from East Asia. Our results show that read assignment to human immunoglobulin isotype genes is very sensitive to the ancestral origin of the genome reference.

The study of mitochondrial DNA (mtDNA) provides invaluable insights into genetic variation, human evolution, and disease mechanisms. However, maintaining a consistent and reliable classification system requires continuous updates. Since Phylotree updates ended in 2016, the accumulation of new haplogroup findings in individual studies has highlighted the critical need for a centralized resource to ensure consistent classifications. To address this gap, we present mitoLEAF, a collaborative, freely accessible, and academically driven repository for mitochondrial phylogenetic analyses. Unlike commercial alternatives that restrict access to their customers through subscription or purchase, mitoLEAF is openly accessible and replicable, ensuring transparency and scientific reproducibility. Hosted as a GitHub repository and supported by an interactive website, mitoLEAF provides an evolving, quality-controlled phylogenetic resource derived from GenBank, EMPOP, and peer-reviewed literature. In this first release, it expands the haplogroup landscape from 5435 to 6409 haplogroups, integrating recent findings and improving phylogenetic accuracy. By excluding known pathogenic variants, mitoLEAF aims to mitigate ethical concerns associated with reporting medically relevant variants. By prioritizing open science over commercial interests, mitoLEAF serves as a vital, community-driven platform for mitochondrial research, fostering collaboration and continuous development.

R-loops are three-stranded RNA and DNA hybrid structures that often occur in the genome and play important roles in a variety of cellular processes from bacteria to mammals. Sequencing methods profiling R-loops genome-wide have revealed that they can form co-transcriptionally at cell type specific genes and associate with specific chromatin states during cell differentiation and reprogramming. However, current computational methods for the prediction of R-loops rely solely on their DNA sequence properties, which precludes detection across cell types, tissues or developmental stages. Here, we conduct a machine learning approach that allows the prediction of mammalian cell type-specific R-loops using sequence information and high-throughput sequencing signals. Our predictive models are induced from human samples and achieve highly accurate predictions, with transcriptomics, DNA features, chromatin accessibility and the active gene body H3K36me3 epigenomic mark being the most informative datasets. We generate de novo virtual R-loop maps that show high concordance with experimental ones and capture cell type specificity. Our approach compares favorably to sequence-based methods and can be generalized to mouse datasets. Based on this, we generate virtual R-loop maps in 51 mammalian systems that are freely accessible to the scientific community.

Single-cell RNA sequencing measures individual cell transcriptomes in a sample. In the past decade, this technology has motivated the development of hundreds of clustering methods. These methods attempt to group cells into populations by leveraging the similarity of their transcriptomes. Because each method relies on specific hypotheses, their predictions can vary drastically. To address this issue, ensemble algorithms detect cell populations by integrating multiple clustering methods, and minimizing the differences of their predictions. While this approach is sensible, it has yet to address some conceptual challenges in single-cell data science; namely, ensemble algorithms have yet to generate clustering results with uncertainty values and multiple resolutions. In this work, we present an original approach to ensemble clustering that addresses these challenges, by describing the differences between clustering results, rather than minimizing them. We present the scEVE algorithm, and we evaluate it on 15 experimental datasets, and up to 1200 synthetic datasets. Our results reveal that scEVE outperforms the state of the art, and addresses both conceptual challenges. We also highlight how biological downstream analyses will benefit from addressing these challenges. We expect that this work will provide an alternative direction for developing single-cell ensemble clustering algorithms.

Reference-based metagenomic profiling requires large genome libraries to maximize detection and minimize false positives. However, as libraries grow, classification accuracy suffers, particularly in k-mer-based tools, as the growing overlap in genomic regions among organisms results in more high-level taxonomic assignments, blunting precision. To address this, we propose sample-tailored minimizer libraries, which improve on the minimizer-lowest common ancestor classification algorithm from the widely used Kraken 2. In this method, an initial filtering step using a large library removes non-resemblance genomes, followed by a refined classification step using a dynamically built smaller minimizer library. This 2-step classification method shows significant performance improvements compared to the state of the art. We develop a new computational tool called Slacken, a distributed and highly scalable platform based on Apache Spark, to implement the 2-step classification method, which improves speed while keeping the cost per sample comparable to Kraken 2. Specifically, in the CAMI2 'strain madness' samples, the fraction of reads classified at species level increased by 3.5×, while for in silico samples, it increased by 2.2×. The 2-step method achieves the sensitivity of large genomic libraries and the specificity of smaller ones, unlocking the true potential of large reference libraries for metagenomic read profiling.

Polyglutamine regions appear in many eukaryotic proteins. Most research on these stretches has focused on humans and primates. We wanted to check whether patterns in their codon usage are shared across a wide taxonomic range. Protein-coding transcripts from 30 eukaryotic model species were searched for stretches of consecutive glutamine codons (CAA/CAG). Most species have higher CAG proportion in longer stretches, except fishes, which either reduced or kept a stable CAG use. CAA codons are located closer to the C-terminal side of the stretches in plants, invertebrates, and tetrapods; fungi showed no bias and fishes showed the opposite. Many tetrapods have codons flanking pure CAG stretches that hint at a mutational control of repeat growth. However, the maximum number of consecutive identical codons within the polyglutamine stretches in most species followed random expectations, with fishes as a main exception. We detected shared patterns in codon usage and position across taxonomically distant species, yet each group retained unique traits. Internal CAA position and external flanking codons both seemed to slow pure CAG expansion. Overall, a mix of random processes and species-specific factors drives how glutamine repeats are shaped and maintained in evolution.