BACKGROUND: Advanced long-read sequencing technologies, such as those from Oxford Nanopore Technologies and Pacific Biosciences, are finding a wide use in de novo genome sequencing projects. However, long reads typically have higher error rates relative to short reads. If left unaddressed, subsequent genome assemblies may exhibit high base error rates that compromise the reliability of downstream analysis. Several specialized error correction tools for genome assemblies have since emerged, employing a range of algorithms and strategies to improve base quality. However, despite these efforts, many genome assembly workflows still produce regions with elevated error rates, such as gaps filled with unpolished or ambiguous bases. To address this, we introduce GoldPolish-Target, a modular targeted sequence polishing pipeline. Coupled with GoldPolish, a linear-time genome assembly algorithm, GoldPolish-Target isolates and polishes user-specified assembly loci, offering a resource-efficient means for polishing targeted regions of draft genomes. RESULTS: Experiments using Drosophila melanogaster and Homo sapiens datasets demonstrate that GoldPolish-Target can reduce insertion/deletion (indel) and mismatch errors by up to 49.2% and 55.4% respectively, achieving base accuracy values upwards of 99.9% (Phred score Q > 30). This polishing accuracy is comparable to the current state-of-the-art, Medaka, while exhibiting up to 27-fold shorter run times and consuming 95% less memory, on average. CONCLUSION: GoldPolish-Target, in contrast to most other polishing tools, offers the ability to target specific regions of a genome assembly for polishing, providing a computationally light-weight and highly scalable solution for base error correction.

Recent deep learning models have shown remarkable performance in image classification. While these deep learning systems are getting closer to practical deployment, the common assumption made about data is that it does not carry any sensitive information. This assumption may not hold for many practical cases, especially in the domain where an individual’s personal information is involved, like healthcare and facial recognition systems. We posit that selectively removing features in this latent space can protect the sensitive information and provide better privacy-utility trade-off. Consequently, we propose DISCO which learns a dynamic and data driven pruning filter to selectively obfuscate sensitive information in the feature space. We propose diverse attack schemes for sensitive inputs & attributes and demonstrate the effectiveness of DISCO against state-of-the-art methods through quantitative and qualitative evaluation. Finally, we also release an evaluation benchmark dataset of 1 million sensitive representations to encourage rigorous exploration of novel attack and defense schemes at https://github.com/splitlearning/InferenceBenchmark.

For models trained on-premise but deployed in a distributed fashion across multiple entities, we demonstrate that minimizing distance correlation between sensitive data such as faces and intermediary representations enables prediction while preventing reconstruction attacks. Leakage (measured using distance correlation between input and intermediate representations) is the risk associated with the reconstruction of raw face data from intermediary representations that are communicated in a distributed setting. We demonstrate on face datasets that our method is resilient to reconstruction attacks during distributed inference while maintaining information required to sustain good classification accuracy. We share modular code for performing NoPeek-Infer at http://tiny.cc/nopeek along with corresponding trained models for benchmarking attack techniques.

Indexes are a critical component of data management applications. While tree-like structures (e.g., B-Trees) have been employed to great success, recent work suggests that index structures powered by machine learning models (learned index structures) can achieve low lookup times with a reduced memory footprint. This demonstration showcases CDFShop, a tool to explore and optimize recursive model indexes (RMIs), a type of learned index structure. This demonstration allows audience members to (1) gain an intuition about various tuning parameters of RMIs and why learned index structures can greatly accelerate search, and (2) understand how automatic optimization techniques can be used to explore space/time tradeoffs within the space of RMIs.