Explainable Artificial Intelligence (XAI) techniques hold significant potential for enhancing the causal discovery process, which is crucial for understanding complex systems in areas like healthcare, economics, and artificial intelligence. However, no causal discovery methods currently incorporate explainability into their models to derive the causal graphs. Thus, in this paper we explore this innovative approach, as it offers substantial potential and represents a promising new direction worth investigating. Specifically, we introduce R e X, a causal discovery method that leverages machine learning (ML) models coupled with explainability techniques, specifically Shapley values, to identify and interpret significant causal relationships among variables. Comparative evaluations on synthetic datasets comprising continuous tabular data reveal that R e X outperforms state-of-the-art causal discovery methods across diverse data generation processes, including non-linear and additive noise models. Moreover, R e X was tested on the Sachs single-cell protein-signaling dataset, achieving a precision of 0.952 and recovering key causal relationships with no incorrect edges. Taking together, these results showcase R e X’s effectiveness in accurately recovering true causal structures while minimizing false positive predictions, its robustness across diverse datasets, and its applicability to real-world problems. By combining ML and explainability techniques with causal discovery, R e X bridges the gap between predictive modeling and causal inference, offering an effective tool for understanding complex causal structures. R e X is publicly available at https://github.com/renero/causalgraph .

Predicting the impact of genomic and drug perturbations in cellular function is crucial for understanding gene functions and drug effects, ultimately leading to improved therapies. To this end, Causal Representation Learning (CRL) constitutes one of the most promising approaches, as it aims to identify the latent factors that causally govern biological systems, thus facilitating the prediction of the effect of unseen perturbations. Yet, current CRL methods fail in reconciling their principled latent representations with known biological processes, leading to models that are not interpretable. To address this major issue, in this work we present SENA-discrepancy-VAE, a model based on the recently proposed CRL method discrepancy-VAE, that produces representations where each latent factor can be interpreted as the (linear) combination of the activity of a (learned) set of biological processes. To this extent, we present an encoder, SENA-$\delta$, that efficiently compute and map biological processes' activity levels to the latent causal factors. We show that SENA-discrepancy-VAE achieves predictive performances on unseen combinations of interventions that are comparable with its original, non-interpretable counterpart, while inferring causal latent factors that are biologically meaningful.

MOTIVATION: Single-nucleotide variants (SNVs) are the most common type of genetic variation in the human genome. Accurate and efficient detection of SNVs from next-generation sequencing (NGS) data is essential for various applications in genomics and personalized medicine. However, SNV calling methods usually suffer from high computational complexity and limited accuracy. In this context, there is a need for new methods that overcome these limitations and provide fast reliable results. RESULTS: We present EMVC-2, a novel method for SNV calling from NGS data. EMVC-2 uses a multi-class ensemble classification approach based on the expectation-maximization algorithm that infers at each locus the most likely genotype from multiple labels provided by different learners. The inferred variants are then validated by a decision tree that filters out unlikely ones. We evaluate EMVC-2 on several publicly available real human NGS data for which the set of SNVs is available, and demonstrate that it outperforms state-of-the-art variant callers in terms of accuracy and speed, on average. AVAILABILITY AND IMPLEMENTATION: EMVC-2 is coded in C and Python, and is freely available for download at: https://github.com/guilledufort/EMVC-2. EMVC-2 is also available in Bioconda.

MOTIVATION: The identification of minimal genetic interventions that modulate metabolic processes constitutes one of the most relevant applications of genome-scale metabolic models (GEMs). The concept of Minimal Cut Sets (MCSs) and its extension at the gene level, genetic Minimal Cut Sets (gMCSs), have attracted increasing interest in the field of Systems Biology to address this task. Different computational tools have been developed to calculate MCSs and gMCSs using both commercial and open-source software. RESULTS: Here, we present gMCSpy, an efficient Python package to calculate gMCSs in GEMs using both commercial and non-commercial optimization solvers. We show that gMCSpy substantially overperforms our previous computational tool GMCS, which exclusively relied on commercial software. Moreover, we compared gMCSpy with recently published competing algorithms in the literature, finding significant improvements in both accuracy and computation time. All these advances make gMCSpy an attractive tool for researchers in the field of Systems Biology for different applications in health and biotechnology. AVAILABILITY AND IMPLEMENTATION: The Python package gMCSpy and the data underlying this manuscript can be accessed at: https://github.com/PlanesLab/gMCSpy.

MOTIVATION: Drug-target interaction (DTI) prediction is a relevant but challenging task in the drug repurposing field. In-silico approaches have drawn particular attention as they can reduce associated costs and time commitment of traditional methodologies. Yet, current state-of-the-art methods present several limitations: existing DTI prediction approaches are computationally expensive, thereby hindering the ability to use large networks and exploit available datasets and, the generalization to unseen datasets of DTI prediction methods remains unexplored, which could potentially improve the development processes of DTI inferring approaches in terms of accuracy and robustness. RESULTS: In this work, we introduce GeNNius (Graph Embedding Neural Network Interaction Uncovering System), a Graph Neural Network (GNN)-based method that outperforms state-of-the-art models in terms of both accuracy and time efficiency across a variety of datasets. We also demonstrated its prediction power to uncover new interactions by evaluating not previously known DTIs for each dataset. We further assessed the generalization capability of GeNNius by training and testing it on different datasets, showing that this framework can potentially improve the DTI prediction task by training on large datasets and testing on smaller ones. Finally, we investigated qualitatively the embeddings generated by GeNNius, revealing that the GNN encoder maintains biological information after the graph convolutions while diffusing this information through nodes, eventually distinguishing protein families in the node embedding space. AVAILABILITY AND IMPLEMENTATION: GeNNius code is available at https://github.com/ubioinformat/GeNNius.

Emerging ultra-low coverage single-cell DNA sequencing (scDNA-seq) technologies have enabled high resolution evolutionary studies of copy number aberrations (CNAs) within tumors. While these sequencing technologies are well suited for identifying CNAs due to the uniformity of sequencing coverage, the sparsity of coverage poses challenges for the study of single-nucleotide variants (SNVs). In order to maximize the utility of increasingly available ultra-low coverage scDNA-seq data and obtain a comprehensive understanding of tumor evolution, it is important to also analyze the evolution of SNVs from the same set of tumor cells. We present Phertilizer, a method to infer a clonal tree from ultra-low coverage scDNA-seq data of a tumor. Based on a probabilistic model, our method recursively partitions the data by identifying key evolutionary events in the history of the tumor. We demonstrate the performance of Phertilizer on simulated data as well as on two real datasets, finding that Phertilizer effectively utilizes the copy-number signal inherent in the data to more accurately uncover clonal structure and genotypes compared to previous methods.

MOTIVATION: An important step in the transcriptomic analysis of individual cells involves manually determining the cellular identities. To ease this labor-intensive annotation of cell-types, there has been a growing interest in automated cell annotation, which can be achieved by training classification algorithms on previously annotated datasets. Existing pipelines employ dataset integration methods to remove potential batch effects between source (annotated) and target (unannotated) datasets. However, the integration and classification steps are usually independent of each other and performed by different tools. We propose JIND (joint integration and discrimination for automated single-cell annotation), a neural-network-based framework for automated cell-type identification that performs integration in a space suitably chosen to facilitate cell classification. To account for batch effects, JIND performs a novel asymmetric alignment in which unseen cells are mapped onto the previously learned latent space, avoiding the need of retraining the classification model for new datasets. JIND also learns cell-type-specific confidence thresholds to identify cells that cannot be reliably classified. RESULTS: We show on several batched datasets that the joint approach to integration and classification of JIND outperforms in accuracy existing pipelines, and a smaller fraction of cells is rejected as unlabeled as a result of the cell-specific confidence thresholds. Moreover, we investigate cells misclassified by JIND and provide evidence suggesting that they could be due to outliers in the annotated datasets or errors in the original approach used for annotation of the target batch. AVAILABILITY AND IMPLEMENTATION: Implementation for JIND is available at https://github.com/mohit1997/JIND and the data underlying this article can be accessed at https://doi.org/10.5281/zenodo.6246322. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Unsupervised ensemble learning refers to methods devised for a particular task that combine data provided by decision learners taking into account their reliability, which is usually inferred from the data. Here, the variant calling step of the next generation sequencing technologies is formulated as an unsupervised ensemble classification problem. A variant calling algorithm based on the expectation-maximization algorithm is further proposed that estimates the maximum-a-posteriori decision among a number of classes larger than the number of different labels provided by the learners. Experimental results with real human DNA sequencing data show that the proposed algorithm is competitive compared to state-of-the-art variant callers as GATK, HTSLIB, and Platypus.

We consider lossless compression based on statistical data modeling followed by prediction-based encoding, where an accurate statistical model for the input data leads to substantial improvements in compression. We propose DZip, a general-purpose compressor for sequential data that exploits the well-known modeling capabilities of neural networks (NNs) for prediction, followed by arithmetic coding. DZip uses a novel hybrid architecture based on adaptive and semi-adaptive training. Unlike most NN-based compressors, DZip does not require additional training data and is not restricted to specific data types. The proposed compressor outperforms general-purpose compressors such as Gzip (29% size reduction on average) and 7zip (12% size reduction on average) on a variety of real datasets, achieves near-optimal compression on synthetic datasets, and performs close to specialized compressors for large sequence lengths, without any human input. While the main limitation of NN-based compressors is generally the encoding/decoding speed, we empirically demonstrate that DZip achieves comparable compression ratio to other NN-based compressors while being several times faster. The source code for DZip and links to the datasets are available at https://github.com/mohit1997/Dzip-torch.

MOTIVATION: Nanopore sequencing technologies are rapidly gaining popularity, in part, due to the massive amounts of genomic data they produce in short periods of time (up to 8.5 TB of data in <72 h). To reduce the costs of transmission and storage, efficient compression methods for this type of data are needed. RESULTS: We introduce RENANO, a reference-based lossless data compressor specifically tailored to FASTQ files generated with nanopore sequencing technologies. RENANO improves on its predecessor ENANO, currently the state of the art, by providing a more efficient base call sequence compression component. Two compression algorithms are introduced, corresponding to the following scenarios: (1) a reference genome is available without cost to both the compressor and the decompressor and (2) the reference genome is available only on the compressor side, and a compacted version of the reference is included in the compressed file. We compare the compression performance of RENANO against ENANO on several publicly available nanopore datasets. RENANO improves the base call sequences compression of ENANO by 39.8% in scenario (1), and by 33.5% in scenario (2), on average, over all the datasets. As for total file compression, the average improvements are 12.7% and 10.6%, respectively. We also show that RENANO consistently outperforms the recent general-purpose genomic compressor Genozip. AVAILABILITY AND IMPLEMENTATION: RENANO is freely available for download at: https://github.com/guilledufort/RENANO. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Mass spectrometry (MS) data, used for proteomics and metabolomics analyses, have seen considerable growth in the last years. Aiming at reducing the associated storage costs, dedicated compression algorithms for MS data have been proposed, such as MassComp and MSNumpress. However, these algorithms focus on either lossless or lossy compression, respectively, and do not exploit the additional redundancy existing across scans contained in a single file. We introduce mspack, a compression algorithm for MS data that exploits this additional redundancy and that supports both lossless and lossy compression, as well as the mzML and the legacy mzXML formats. mspack applies several preprocessing lossless transforms and optional lossy transforms with a configurable error, followed by the general purpose compressors gzip or bsc to achieve a higher compression ratio. RESULTS: We tested mspack on several datasets generated by commonly used MS instruments. When used with the bsc compression backend, mspack achieves on average 76% smaller file sizes for lossless compression and 94% smaller file sizes for lossy compression, as compared with the original files. Lossless mspack achieves 10-60% lower file sizes than MassComp, and lossy mspack compresses 36-60% better than the lossy MSNumpress, for the same error, while exhibiting comparable accuracy and running time. AVAILABILITY AND IMPLEMENTATION: mspack is implemented in C++ and freely available at https://github.com/fhanau/mspack under the Apache license. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

We consider lossless compression based on statistical data modeling followed by prediction-based encoding, where an accurate statistical model for the input data leads to substantial improvements in compression. We propose DZip, a general-purpose compressor for sequential data that exploits the well-known modeling capabilities of neural networks (NNs) for prediction, followed by arithmetic coding. DZip uses a novel hybrid architecture based on adaptive and semi-adaptive training. Unlike most NN based compressors, DZip does not require additional training data and is not restricted to specific data types, only needing the alphabet size of the input data. The proposed compressor outperforms general-purpose compressors such as Gzip (on average 26% reduction) on a variety of real datasets, achieves near-optimal compression on synthetic datasets, and performs close to specialized compressors for large sequence lengths, without any human input. The main limitation of DZip in its current implementation is the encoding/decoding time, which limits its practicality. Nevertheless, the results showcase the potential of developing improved general-purpose compressors based on neural networks and hybrid modeling.

MOTIVATION: The amount of genomic data generated globally is seeing explosive growth, leading to increasing needs for processing, storage and transmission resources, which motivates the development of efficient compression tools for these data. Work so far has focused mainly on the compression of data generated by short-read technologies. However, nanopore sequencing technologies are rapidly gaining popularity due to the advantages offered by the large increase in the average size of the produced reads, the reduction in their cost and the portability of the sequencing technology. We present ENANO (Encoder for NANOpore), a novel lossless compression algorithm especially designed for nanopore sequencing FASTQ files. RESULTS: The main focus of ENANO is on the compression of the quality scores, as they dominate the size of the compressed file. ENANO offers two modes, Maximum Compression and Fast (default), which trade-off compression efficiency and speed. We tested ENANO, the current state-of-the-art compressor SPRING and the general compressor pigz on several publicly available nanopore datasets. The results show that the proposed algorithm consistently achieves the best compression performance (in both modes) on every considered nanopore dataset, with an average improvement over pigz and SPRING of >24.7% and 6.3%, respectively. In addition, in terms of encoding and decoding speeds, ENANO is 2.9× and 1.7× times faster than SPRING, respectively, with memory consumption up to 0.2 GB. AVAILABILITY AND IMPLEMENTATION: ENANO is freely available for download at: https://github.com/guilledufort/EnanoFASTQ. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Sequencing data are often summarized at different annotation levels for further analysis, generally using the general feature format (GFF) or its descendants, gene transfer format (GTF) and GFF3. Existing utilities for accessing these files, like gffutils and gffread, do not focus on reducing the storage space, significantly increasing it in some cases. We propose GPress, a framework for querying GFF files in a compressed form. GPress can also incorporate and compress expression files from both bulk and single-cell RNA-Seq experiments, supporting simultaneous queries on both the GFF and expression files. In brief, GPress applies transformations to the data which are then compressed with the general lossless compressor BSC. To support queries, GPress compresses the data in blocks and creates several index tables for fast retrieval. RESULTS: We tested GPress on several GFF files of different organisms, and showed that it achieves on average a 61% reduction in size with respect to gzip (the current de facto compressor for GFF files) while being able to retrieve all annotations for a given identifier or a range of coordinates in a few seconds (when run in a common laptop). In contrast, gffutils provides faster retrieval but doubles the size of the GFF files. When additionally linking an expression file, we show that GPress can reduce its size by more than 68% when compared to gzip (for both bulk and single-cell RNA-Seq experiments), while still retrieving the information within seconds. Finally, applying BSC to the data streams generated by GPress instead of to the original file shows a size reduction of more than 44% on average. AVAILABILITY AND IMPLEMENTATION: GPress is freely available at https://github.com/qm2/gpress. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: In an effort to provide a response to the ever-expanding generation of genomic data, the International Organization for Standardization (ISO) is designing a new solution for the representation, compression and management of genomic sequencing data: the Moving Picture Experts Group (MPEG)-G standard. This paper discusses the first implementation of an MPEG-G compliant entropy codec: GABAC. GABAC combines proven coding technologies, such as context-adaptive binary arithmetic coding, binarization schemes and transformations, into a straightforward solution for the compression of sequencing data. RESULTS: We demonstrate that GABAC outperforms well-established (entropy) codecs in a significant set of cases and thus can serve as an extension for existing genomic compression solutions, such as CRAM. AVAILABILITY AND IMPLEMENTATION: The GABAC library is written in C++. We also provide a command line application which exercises all features provided by the library. GABAC can be downloaded from https://github.com/mitogen/gabac. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Variants identified by current genomic analysis pipelines contain many incorrectly called variants. These can be potentially eliminated by applying state-of-the-art filtering tools, such as Variant Quality Score Recalibration (VQSR) or Hard Filtering (HF). However, these methods are very user-dependent and fail to run in some cases. We propose VEF, a variant filtering tool based on decision tree ensemble methods that overcomes the main drawbacks of VQSR and HF. Contrary to these methods, we treat filtering as a supervised learning problem, using variant call data with known 'true' variants, i.e. gold standard, for training. Once trained, VEF can be directly applied to filter the variants contained in a given Variants Call Format (VCF) file (we consider training and testing VCF files generated with the same tools, as we assume they will share feature characteristics). RESULTS: For the analysis, we used whole genome sequencing (WGS) Human datasets for which the gold standards are available. We show on these data that the proposed filtering tool VEF consistently outperforms VQSR and HF. In addition, we show that VEF generalizes well even when some features have missing values, when the training and testing datasets differ in coverage, and when sequencing pipelines other than GATK are used. Finally, since the training needs to be performed only once, there is a significant saving in running time when compared with VQSR (4 versus 50 min approximately for filtering the single nucleotide polymorphisms of a WGS Human sample). AVAILABILITY AND IMPLEMENTATION: Code and scripts available at: github.com/ChuanyiZ/vef. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Sequential data is being generated at an unprecedented pace in various forms, including text and genomic data. This creates the need for efficient compression mechanisms to enable better storage, transmission and processing of such data. To solve this problem, many of the existing compressors attempt to learn models for the data and perform prediction-based compression. Since neural networks are known as universal function approximators with the capability to learn arbitrarily complex mappings, and in practice show excellent performance in prediction tasks, we explore and devise methods to compress sequential data using neural network predictors. We combine recurrent neural network predictors with an arithmetic coder and losslessly compress a variety of synthetic, text and genomic datasets. The proposed compressor outperforms Gzip on the real datasets and achieves near-optimal compression for the synthetic datasets. The results also help understand why and where neural networks are good alternatives for traditional finite context models.

As genome sequencing continues to become more cost-effective and affordable, more raw and aligned genomic files are expected to be generated in future years. In addition, due to the increase in the throughput of sequencing machines, the size of these files is significantly growing. In particular, aligned files (e.g., SAM/BAM) are used for further processing of the data, and hence efficient representation of these files is a pressing need. In this work we present AliCo, a new compression method tailored to the aligned data represented in the SAM format. We demonstrate through simulations on existing datasets that AliCo outperforms in compression ratio, on average, the state-of-the-art compressors for SAM files, achieving more than 85% reduction in size when operating in its lossless mode. AliCo also supports a variety of modes for lossy compression of the quality scores, including for the first time the recently proposed lossy compressor CALQ, which uses information from the aligned reads to adjust the level of quantization for each location of the genome (achieving more than 10× compression gains in high-coverage datasets). AliCo also supports optional compression of the reference sequence used for compression, hence guaranteeing exact reconstruction of the compressed data. Finally, AliCo allows to stream the data as it is being compressed, as well as to decompress the data as it is being received, potentially providing significant time savings. AliCo can be accessed at: https://github.com/iochoa/alico.

MOTIVATION: High-Throughput Sequencing technologies produce huge amounts of data in the form of short genomic reads, associated quality values and read identifiers. Because of the significant structure present in these FASTQ datasets, general-purpose compressors are unable to completely exploit much of the inherent redundancy. Although there has been a lot of work on designing FASTQ compressors, most of them lack in support of one or more crucial properties, such as support for variable length reads, scalability to high coverage datasets, pairing-preserving compression and lossless compression. RESULTS: In this work, we propose SPRING, a reference-free compressor for FASTQ files. SPRING supports a wide variety of compression modes and features, including lossless compression, pairing-preserving compression, lossy compression of quality values, long read compression and random access. SPRING achieves substantially better compression than existing tools, for example, SPRING compresses 195 GB of 25× whole genome human FASTQ from Illumina's NovaSeq sequencer to less than 7 GB, around 1.6× smaller than previous state-of-the-art FASTQ compressors. SPRING achieves this improvement while using comparable computational resources. AVAILABILITY AND IMPLEMENTATION: SPRING can be downloaded from https://github.com/shubhamchandak94/SPRING. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Mass Spectrometry (MS) is a widely used technique in biology research, and has become key in proteomics and metabolomics analyses. As a result, the amount of MS data has significantly increased in recent years. For example, the MS repository MassIVE contains more than 123TB of data. Somehow surprisingly, these data are stored uncompressed, hence incurring a significant storage cost. Efficient representation of these data is therefore paramount to lessen the burden of storage and facilitate its dissemination. RESULTS: We present MassComp, a lossless compressor optimized for the numerical (m/z)-intensity pairs that account for most of the MS data. We tested MassComp on several MS data and show that it delivers on average a 46% reduction on the size of the numerical data, and up to 89%. These results correspond to an average improvement of more than 27% when compared to the general compressor gzip and of 40% when compared to the state-of-the-art numerical compressor FPC. When tested on entire files retrieved from the MassIVE repository, MassComp achieves on average a 59% size reduction. MassComp is written in C++ and freely available at https://github.com/iochoa/MassComp . CONCLUSIONS: The compression performance of MassComp demonstrates its potential to significantly reduce the footprint of MS data, and shows the benefits of designing specialized compression algorithms tailored to MS data. MassComp is an addition to the family of omics compression algorithms designed to lessen the storage burden and facilitate the exchange and dissemination of omics data.

Bioinformatics (2015) 31(19), 3122–3129 The authors of the above article wish to inform readers that a post-production correction has been made to add missing funding information: NIH grant U01 CA198943.

Motivation: DNA methylation is one of the most important epigenetic mechanisms in cells that exhibits a significant role in controlling gene expressions. Abnormal methylation patterns have been associated with cancer, imprinting disorders and repeat-instability diseases. As inexpensive bisulfite sequencing approaches have led to significant efforts in acquiring methylation data, problems of data storage and management have become increasingly important. The de facto compression method for methylation data is gzip, which is a general purpose compression algorithm that does not cater to the special format of methylation files. We propose METHCOMP, a new compression scheme tailor-made for bedMethyl files, which supports random access. Results: We tested the METHCOMP algorithm on 24 bedMethyl files retrieved from four randomly selected ENCODE assays. Our findings reveal that METHCOMP offers an average compression ratio improvement over gzip of up to 7.5x. As an example, METHCOMP compresses a 48 GB file to only 0.9 GB, which corresponds to a 98% reduction in size. Availability and implementation: METHCOMP is freely available at https://github.com/jianhao2016/METHCOMP. Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: The affordability of DNA sequencing has led to the generation of unprecedented volumes of raw sequencing data. These data must be stored, processed and transmitted, which poses significant challenges. To facilitate this effort, we introduce FaStore, a specialized compressor for FASTQ files. FaStore does not use any reference sequences for compression and permits the user to choose from several lossy modes to improve the overall compression ratio, depending on the specific needs. Results: FaStore in the lossless mode achieves a significant improvement in compression ratio with respect to previously proposed algorithms. We perform an analysis on the effect that the different lossy modes have on variant calling, the most widely used application for clinical decision making, especially important in the era of precision medicine. We show that lossy compression can offer significant compression gains, while preserving the essential genomic information and without affecting the variant calling performance. Availability and implementation: FaStore can be downloaded from https://github.com/refresh-bio/FaStore. Supplementary information: Supplementary data are available at Bioinformatics online.

The affordability of DNA sequencing has led to unprecedented volumes of genomic data. These data must be stored, processed, and analyzed. The most popular format for genomic data is the SAM format, which contains information such as alignment, quality values, etc. These files are large (on the order of terabytes), which necessitates compression. In this work we propose a new reference-based compressor for SAM files, which can accommodate different levels of compression, based on the specific needs of the user. In particular, the proposed compressor GeneComp allows the user to perform lossy compression of the quality scores, which have been proven to occupy more than half of the compressed file (when losslessly compressed). We show that the proposed compressor GeneComp overall achieves better compression ratios than previously proposed algorithms when working on lossless mode.

Recent advancements in sequencing technology have led to a drastic reduction in genome sequencing costs. This development has generated an unprecedented amount of data that must be stored, processed, and communicated. To facilitate this effort, compression of genomic files has been proposed. Specifically, lossy compression of quality scores is emerging as a natural candidate for reducing the growing costs of storage. A main goal of performing DNA sequencing in population studies and clinical settings is to identify genetic variation. Though the field agrees that smaller files are advantageous, the cost of lossy compression, in terms of variant discovery, is unclear.Bioinformatic algorithms to identify SNPs and INDELs use base quality score information; here, we evaluate the effect of lossy compression of quality scores on SNP and INDEL detection. Specifically, we investigate how the output of the variant caller when using the original data differs from that obtained when quality scores are replaced by those generated by a lossy compressor. Using gold standard genomic datasets and simulated data, we are able to analyze how accurate the output of the variant calling is, both for the original data and that previously lossily compressed. We show that lossy compression can significantly alleviate the storage while maintaining variant calling performance comparable to that with the original data. Further, in some cases lossy compression can lead to variant calling performance that is superior to that using the original file. We envisage our findings and framework serving as a benchmark in future development and analyses of lossy genomic data compressors.

Massive amounts of sequencing data are being generated thanks to advances in sequencing technology and a dramatic drop in the sequencing cost. Storing and sharing this large data has become a major bottleneck in the discovery and analysis of genetic variants that are used for medical inference. As such, lossless compression of this data has been proposed. Of the compressed data, more than 70% correspond to quality scores, which indicate the sequencing machine reliability when calling a particular basepair. Thus, to further improve the compression performance, lossy compression of quality scores is emerging as the natural candidate. Since the data is used for genetic variants discovery, lossy compressors for quality scores are analyzed in terms of their rate-distortion performance, as well as their effect on the variant callers. Previously proposed algorithms do not do well under all performance metrics, and are hence unsuitable for certain applications. In this work we propose a new lossy compressor that first performs a clustering step, by assuming all the quality scores sequences come from a mixture of Markov models. Then, it performs quantization of the quality scores based on the Markov models. Each quantizer targets a specific distortion to optimize for the overall rate-distortion performance. Finally, the quantized values are compressed by an entropy encoder. We demonstrate that the proposed lossy compressor outperforms the previously proposed methods under all analyzed distortion metrics. This suggests that the effect that the proposed algorithm will have on any downstream application will likely be less noticeable than that of previously proposed lossy compressors. Moreover, we analyze how the proposed lossy compressor affects Single Nucleotide Polymorphism (SNP) calling, and show that the variability introduced on the calls is considerably smaller than the variability that exists between different methodologies for SNP calling.

Massive amounts of sequencing data are being generated thanks to advances in sequencing technology and a dramatic drop in the sequencing cost. Much of the raw data are comprised of nucleotides and the corresponding quality scores that indicate their reliability. The latter are more difficult to compress and are themselves noisy. Lossless and lossy compression of the quality scores has recently been proposed to alleviate the storage costs, but reducing the noise in the quality scores has remained largely unexplored. This raw data is processed in order to identify variants; these genetic variants are used in important applications, such as medical decision making. Thus improving the performance of the variant calling by reducing the noise contained in the quality scores is important. We propose a denoising scheme that reduces the noise of the quality scores and we demonstrate improved inference with this denoised data. Specifically, we show that replacing the quality scores with those generated by the proposed denoiser results in more accurate variant calling in general. Moreover, a consequence of the denoising is that the entropy of the produced quality scores is smaller, and thus significant compression can be achieved with respect to lossless compression of the original quality scores. We expect our results to provide a baseline for future research in denoising of quality scores. The code used in this work as well as a Supplement with all the results are available at http://web.stanford.edu/~iochoa/DCCdenoiser_CodeAndSupplement.zip.

Massive amounts of sequencing data are being generated thanks to advances in sequencing technology and a dramatic drop in the sequencing cost. Much of the data are comprised of nucleotides and the corresponding quality scores that indicate their reliability. The latter are more difficult to compress and are themselves noisy. As a result, lossy compression of the quality scores has recently been proposed to alleviate the storage costs. Further, it has been shown that lossy compression, at some specific rates, can achieve a performance on variant calling similar to that achieved with the lossless compressed data. We propose CROMqs, a new lossy compressor for the quality scores with the property of "infinitesimal successive refinability". This property allows the decoder to decompress the data iteratively without the need of agreeing with the encoder on a specific rate prior to compression. This characteristic is particularly amenable in practice, as in most cases the appropriate rate at which the lossy compressor should operate can not be established prior to compression. Further, this property can be of interest in scenarios involving streaming of genomic data. CROMqs is the first infinitesimal successive refinement lossy compressor for the quality scores in the literature, and we show that it obtains a comparable rate-distortion performance to previously proposed algorithms. Moreover, we also show that CROMqs achieves a comparable performance on variant calling to that of the lossless compressed data.

MOTIVATION: Data compression is crucial in effective handling of genomic data. Among several recently published algorithms, ERGC seems to be surprisingly good, easily beating all of the competitors. RESULTS: We evaluated ERGC and the previously proposed algorithms GDC and iDoComp, which are the ones used in the original paper for comparison, on a wide data set including 12 assemblies of human genome (instead of only four of them in the original paper). ERGC wins only when one of the genomes (referential or target) contains mixed-cased letters (which is the case for only the two Korean genomes). In all other cases ERGC is on average an order of magnitude worse than GDC and iDoComp. CONTACT: [email protected], [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The dramatic decrease in the cost of sequencing has resulted in the generation of huge amounts of genomic data, as evidenced by projects such as the UK10K and the Million Veteran Project, with the number of sequenced genomes ranging in the order of 10 K to 1 M. Due to the large redundancies among genomic sequences of individuals from the same species, most of the medical research deals with the variants in the sequences as compared with a reference sequence, rather than with the complete genomic sequences. Consequently, millions of genomes represented as variants are stored in databases. These databases are constantly updated and queried to extract information such as the common variants among individuals or groups of individuals. Previous algorithms for compression of this type of databases lack efficient random access capabilities, rendering querying the database for particular variants and/or individuals extremely inefficient, to the point where compression is often relinquished altogether. RESULTS: We present a new algorithm for this task, called GTRAC, that achieves significant compression ratios while allowing fast random access over the compressed database. For example, GTRAC is able to compress a Homo sapiens dataset containing 1092 samples in 1.1 GB (compression ratio of 160), while allowing for decompression of specific samples in less than a second and decompression of specific variants in 17 ms. GTRAC uses and adapts techniques from information theory, such as a specialized Lempel-Ziv compressor, and tailored succinct data structures. AVAILABILITY AND IMPLEMENTATION: The GTRAC algorithm is available for download at: https://github.com/kedartatwawadi/GTRAC CONTACT: : [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Recent advancements in sequencing technology have led to a drastic reduction in the cost of sequencing a genome. This has generated an unprecedented amount of genomic data that must be stored, processed and transmitted. To facilitate this effort, we propose a new lossy compressor for the quality values presented in genomic data files (e.g. FASTQ and SAM files), which comprise roughly half of the storage space (in the uncompressed domain). Lossy compression allows for compression of data beyond its lossless limit. RESULTS: The proposed algorithm QVZ exhibits better rate-distortion performance than the previously proposed algorithms, for several distortion metrics and for the lossless case. Moreover, it allows the user to define any quasi-convex distortion function to be minimized, a feature not supported by the previous algorithms. Finally, we show that QVZ-compressed data exhibit better performance in the genotyping than data compressed with previously proposed algorithms, in the sense that for a similar rate, a genotyping closer to that achieved with the original quality values is obtained. AVAILABILITY AND IMPLEMENTATION: QVZ is written in C and can be downloaded from https://github.com/mikelhernaez/qvz. CONTACT: [email protected] or [email protected] or [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: With the release of the latest next-generation sequencing (NGS) machine, the HiSeq X by Illumina, the cost of sequencing a Human has dropped to a mere $4000. Thus we are approaching a milestone in the sequencing history, known as the $1000 genome era, where the sequencing of individuals is affordable, opening the doors to effective personalized medicine. Massive generation of genomic data, including assembled genomes, is expected in the following years. There is crucial need for compression of genomes guaranteed of performing well simultaneously on different species, from simple bacteria to humans, which will ease their transmission, dissemination and analysis. Further, most of the new genomes to be compressed will correspond to individuals of a species from which a reference already exists on the database. Thus, it is natural to propose compression schemes that assume and exploit the availability of such references. RESULTS: We propose iDoComp, a compressor of assembled genomes presented in FASTA format that compresses an individual genome using a reference genome for both the compression and the decompression. In terms of compression efficiency, iDoComp outperforms previously proposed algorithms in most of the studied cases, with comparable or better running time. For example, we observe compression gains of up to 60% in several cases, including H.sapiens data, when comparing with the best compression performance among the previously proposed algorithms. AVAILABILITY: iDoComp is written in C and can be downloaded from: http://www.stanford.edu/~iochoa/iDoComp.html (We also provide a full explanation on how to run the program and an example with all the necessary files to run it.).

We consider compression of sequences in a database so that similarity queries can be performed efficiently in the compressed domain. The fundamental limits for this problem setting, which characterize the trade off between compression rate and reliability of the answers to the queries, have been characterized in past work. However, how to approach these limits in practice has remained largely unexplored. Recently, we proposed a scheme for this task that is based on existing lossy compression algorithms, for the general case where the similarity measure satisfies a triangle inequality. Although it was shown that it achieves the fundamental limits for some cases, it is suboptimal in general. In this paper we propose a new scheme that also uses lossy compression algorithms as a building block, but with a carefully chosen distortion measure that is different than the one defining the similarity between sequences. The new scheme significantly improves the compression rate compared to the previously proposed scheme in many cases. For example, for binary sources and Hamming similarity measure, simulation results show a compression rate close to the fundamental limit, and an improvement over the previously proposed scheme of up to 55% (for the same reliability). The results shed light on the fact that compression for similarity identification is inherently different than classical lossy compression.

BACKGROUND: Next Generation Sequencing technologies have revolutionized many fields in biology by reducing the time and cost required for sequencing. As a result, large amounts of sequencing data are being generated. A typical sequencing data file may occupy tens or even hundreds of gigabytes of disk space, prohibitively large for many users. This data consists of both the nucleotide sequences and per-base quality scores that indicate the level of confidence in the readout of these sequences. Quality scores account for about half of the required disk space in the commonly used FASTQ format (before compression), and therefore the compression of the quality scores can significantly reduce storage requirements and speed up analysis and transmission of sequencing data. RESULTS: In this paper, we present a new scheme for the lossy compression of the quality scores, to address the problem of storage. Our framework allows the user to specify the rate (bits per quality score) prior to compression, independent of the data to be compressed. Our algorithm can work at any rate, unlike other lossy compression algorithms. We envisage our algorithm as being part of a more general compression scheme that works with the entire FASTQ file. Numerical experiments show that we can achieve a better mean squared error (MSE) for small rates (bits per quality score) than other lossy compression schemes. For the organism PhiX, whose assembled genome is known and assumed to be correct, we show that it is possible to achieve a significant reduction in size with little compromise in performance on downstream applications (e.g., alignment). CONCLUSIONS: QualComp is an open source software package, written in C and freely available for download at https://sourceforge.net/projects/qualcomp.

DNA sequencing technology has advanced to a point where storage is becoming the central bottleneck in the acquisition and mining of more data. Large amounts of data are vital for genomics research, and generic compression tools, while viable, cannot offer the same savings as approaches tuned to inherent biological properties. We propose an algorithm to compress a target genome given a known reference genome. The proposed algorithm first generates a mapping from the reference to the target genome, and then compresses this mapping with an entropy coder. As an illustration of the performance: applying our algorithm to James Watson's genome with hg18 as a reference, we are able to reduce the 2991 megabyte (MB) genome down to 6.99 MB, while Gzip compresses it to 834.8 MB.