Deciphering RNA splicing logic with interpretable machine learning.

Machine learning methods, particularly neural networks trained on large datasets, are transforming how scientists approach scientific discovery and experimental design. However, current state-of-the-art neural networks are limited by their uninterpretability: Despite their excellent accuracy, they cannot describe how they arrived at their predictions. Here, using an "interpretable-by-design" approach, we present a neural network model that provides insights into RNA splicing, a fundamental process in the transfer of genomic information into functional biochemical products. Although we designed our model to emphasize interpretability, its predictive accuracy is on par with state-of-the-art models. To demonstrate the model's interpretability, we introduce a visualization that, for any given exon, allows us to trace and quantify the entire decision process from input sequence to output splicing prediction. Importantly, the model revealed uncharacterized components of the splicing logic, which we experimentally validated. This study highlights how interpretable machine learning can advance scientific discovery.

Splicing at the phase-separated nuclear speckle interface: a model.

Phase-separated membraneless bodies play important roles in nucleic acid biology. While current models for the roles of phase separation largely focus on the compartmentalization of constituent proteins, we reason that other properties of phase separation may play functional roles. Specifically, we propose that interfaces of phase-separated membraneless bodies could have functional roles in spatially organizing biochemical reactions. Here we propose such a model for the nuclear speckle, a membraneless body implicated in RNA splicing. In our model, sequence-dependent RNA positioning along the nuclear speckle interface coordinates RNA splicing. Our model asserts that exons are preferentially sequestered into nuclear speckles through binding by SR proteins, while introns are excluded through binding by nucleoplasmic hnRNP proteins. As a result, splice sites at exon-intron boundaries are preferentially positioned at nuclear speckle interfaces. This positioning exposes splice sites to interface-localized spliceosomes, enabling the subsequent splicing reaction. Our model provides a simple mechanism that seamlessly explains much of the complex logic of splicing. This logic includes experimental results such as the antagonistic duality between splicing factors, the position dependence of splicing sequence motifs, and the collective contribution of many motifs to splicing decisions. Similar functional roles for phase-separated interfaces may exist for other membraneless bodies.

Uncovering Distinct Peptide Charging Behaviors in Electrospray Ionization Mass Spectrometry Using a Large-Scale Dataset.

Electrospray ionization is a powerful and prevalent technique used to ionize analytes in mass spectrometry. The distribution of charges that an analyte receives (charge state distribution, CSD) is an important consideration for interpreting mass spectra. However, due to an incomplete understanding of the ionization mechanism, the analyte properties that influence CSDs are not fully understood. Here, we employ a machine learning-based approach and analyze CSDs of hundreds of thousands of peptides. Interestingly, half of the peptides exhibit charges that differ from what one would naively expect (the number of basic sites). We find that these peptides can be classified into two regimes (undercharging and overcharging) and that these two regimes display markedly different charging characteristics. Notably, peptides in the overcharging regime show minimal dependence on basic site count, and more generally, the two regimes exhibit distinct sequence determinants. These findings highlight the rich ionization behavior of peptides and the potential of CSDs for enhancing peptide identification.

Dynamics of RNA localization to nuclear speckles are connected to splicing efficiency.

Nuclear speckles, a type of membraneless nuclear organelle in higher eukaryotic cells, play a vital role in gene expression regulation. Using the reverse transcription-based RNA-binding protein binding sites sequencing (ARTR-seq) method, we study human transcripts associated with nuclear speckles. We identify three gene groups whose transcripts demonstrate different speckle localization properties and dynamics: stably enriched in nuclear speckles, transiently enriched in speckles at the pre-mRNA stage, and not enriched in speckles. Specifically, we find that stably-enriched transcripts contain inefficiently spliced introns. We show that nuclear speckles specifically facilitate splicing of speckle-enriched transcripts. We further reveal RNA sequence features contributing to transcript speckle localization, underscoring a tight interplay between genome organization, RNA cis-elements, and transcript speckle enrichment, and connecting transcript speckle localization with splicing efficiency. Finally, we show that speckles can act as hubs for the regulated retention of introns during cellular stress. Collectively, our data highlight a role of nuclear speckles in both co- and post-transcriptional splicing regulation.

RNA molecules display distinctive organization at nuclear speckles.

RNA molecules often play critical roles in assisting the formation of membraneless organelles in eukaryotic cells. Yet, little is known about the organization of RNAs within membraneless organelles. Here, using super-resolution imaging and nuclear speckles as a model system, we demonstrate that different sequence domains of RNA transcripts exhibit differential spatial distributions within speckles. Specifically, we image transcripts containing a region enriched in binding motifs of serine/arginine-rich (SR) proteins and another region enriched in binding motifs of heterogeneous nuclear ribonucleoproteins (hnRNPs). We show that these transcripts localize to the outer shell of speckles, with the SR motif-rich region localizing closer to the speckle center relative to the hnRNP motif-rich region. Further, we identify that this intra-speckle RNA organization is driven by the strength of RNA-protein interactions inside and outside speckles. Our results hint at novel functional roles of nuclear speckles and likely other membraneless organelles in organizing RNA substrates for biochemical reactions.

A high-throughput approach reveals distinct peptide charging behaviors in electrospray ionization mass spectrometry.

Electrospray ionization is a powerful and prevalent technique used to ionize analytes in mass spectrometry. The distribution of charges that an analyte receives (charge state distribution, CSD) is an important consideration for interpreting mass spectra. However, due to an incomplete understanding of the ionization mechanism, the analyte properties that influence CSDs are not fully understood. Here, we employ a machine learning-based high-throughput approach and analyze CSDs of hundreds of thousands of peptides. Interestingly, half of the peptides exhibit charges that differ from what one would naively expect (number of basic sites). We find that these peptides can be classified into two regimes-undercharging and overcharging-and that these two regimes display markedly different charging characteristics. Strikingly, peptides in the overcharging regime show minimal dependence on basic site count, and more generally, the two regimes exhibit distinct sequence determinants. These findings highlight the rich ionization behavior of peptides and the potential of CSDs for enhancing peptide identification.

A recurrent COL6A1 pseudoexon insertion causes muscular dystrophy and is effectively targeted by splice-correction therapies.

The clinical application of advanced next-generation sequencing technologies is increasingly uncovering novel classes of mutations that may serve as potential targets for precision medicine therapeutics. Here, we show that a deep intronic splice defect in the COL6A1 gene, originally discovered by applying muscle RNA sequencing in patients with clinical findings of collagen VI-related dystrophy (COL6-RD), inserts an in-frame pseudoexon into COL6A1 mRNA, encodes a mutant collagen α1(VI) protein that exerts a dominant-negative effect on collagen VI matrix assembly, and provides a unique opportunity for splice-correction approaches aimed at restoring normal gene expression. Using splice-modulating antisense oligomers, we efficiently skipped the pseudoexon in patient-derived fibroblast cultures and restored a wild-type matrix. Similarly, we used CRISPR/Cas9 to precisely delete an intronic sequence containing the pseudoexon and efficiently abolish its inclusion while preserving wild-type splicing. Considering that this splice defect is emerging as one of the single most frequent mutations in COL6-RD, the design of specific and effective splice-correction therapies offers a promising path for clinical translation.