We present BootComp, a novel framework based on text-to-image diffusion models for controllable human image generation with multiple reference garments. Here, the main bottleneck is data acquisition for training: collecting a large-scale dataset of high-quality reference garment images per human subject is quite challenging, i.e., ideally, one needs to manually gather every single garment photograph worn by each human. To address this, we propose a data generation pipeline to construct a large synthetic dataset, consisting of human and multiple-garment pairs, by introducing a model to extract any reference garment images from each human image. To ensure data quality, we also propose a filtering strategy to remove undesirable generated data based on measuring perceptual similarities between the garment presented in human image and extracted garment. Finally, by utilizing the constructed synthetic dataset, we train a diffusion model having two parallel denoising paths that use multiple garment images as conditions to generate human images while preserving their fine-grained details. We further show the wide-applicability of our framework by adapting it to different types of reference-based generation in the fashion domain, including virtual try-on, and controllable human image generation with other conditions, e.g., pose, face, etc.

A choice of optimization objective is immensely pivotal in the design of a recommender system as it affects the general modeling process of a user's intent from previous interactions. Existing approaches mainly adhere to three categories of loss functions: pairwise, pointwise, and setwise loss functions. Despite their effectiveness, a critical and common drawback of such objectives is viewing the next observed item as a unique positive while considering all remaining items equally negative. Such a binary label assignment is generally limited to assuring a higher recommendation score of the positive item, neglecting potential structures induced by varying preferences between other unobserved items. To alleviate this issue, we propose a novel method that extends original objectives to explicitly leverage the different levels of preferences as relative orders between their scores. Finally, we demonstrate the superior performance of our method compared to baseline objectives.

Statistical analysis of ultrahigh-dimensional omics scale data has long depended on univariate hypothesis testing. With growing data features and samples, the obvious next step is to establish multivariable association analysis as a routine method to describe genotype-phenotype association. Here we present ParProx, a state-of-the-art implementation to optimize overlapping and non-overlapping group lasso regression models for time-to-event and classification analysis, with selection of variables grouped by biological priors. ParProx enables multivariable model fitting for ultrahigh-dimensional data within an architecture for parallel or distributed computing via latent variable group representation. It thereby aims to produce interpretable regression models consistent with known biological relationships among independent variables, a property often explored post hoc, not during model estimation. Simulation studies clearly demonstrate the scalability of ParProx with graphics processing units in comparison to existing implementations. We illustrate the tool using three different omics data sets featuring moderate to large numbers of variables, where we use genomic regions and biological pathways as variable groups, rendering the selected independent variables directly interpretable with respect to those groups. ParProx is applicable to a wide range of studies using ultrahigh-dimensional omics data, from genome-wide association analysis to multi-omics studies where model estimation is computationally intractable with existing implementation.

Summary: Data visualization is often regarded as a post hoc step for verifying statistically significant results in the analysis of high-throughput datasets. This common practice leaves a large amount of raw data behind, from which more information can be extracted. However, existing solutions do not provide capabilities to explore large-scale raw datasets using biologically sensible queries, nor do they allow user interaction based real-time customization of graphics. To address these drawbacks, we have designed an open-source, web-based tool called Systems-Level Interactive Data Exploration, or SLIDE to visualize large-scale -omics data interactively. SLIDE's interface makes it easier for scientists to explore quantitative expression data in multiple resolutions in a single screen. Availability and implementation: SLIDE is publicly available under BSD license both as an online version as well as a stand-alone version at https://github.com/soumitag/SLIDE. Supplementary information: Supplementary data are available at Bioinformatics online.

This paper studies the problem of parameter learning in graphical models having latent variables, where the standard approach is the expectation maximization algorithm alternating expectation (E) and maximization (M) steps. However, both E and M steps are computationally intractable for high dimensional data, while the substitution of one step to a faster surrogate for combating against intractability can often cause failure in convergence. To tackle the issue, the Contrastive Divergence (CD) learning scheme has been popularly used in the deep learning community, where it runs the mean-field approximation in E step and a few cycles of Markov Chains (MC) in M step. In this paper, we analyze a variant of CD, called Adiabatic Persistent Contrastive Divergence (APCD), which runs a few cycles of MCs in both E and M steps. Using multi-time-scale stochastic approximation theory, we prove that APCD converges to a correct optimum, where the standard CD is impossible to have such a guarantee due to the mean-field approximation gap in E step. Despite of such stronger theoretical guarantee of APCD, its possible drawback is on slow mixing on E step for practical purposes. To address the issue, we also design a hybrid approach applying both mean-field and MC approximations in E step, where it outperforms the standard mean-field-based CD in our experiments on real-world datasets.

MOTIVATION: Accurate cross-sample peak alignment and reliable intensity normalization is a critical step for robust quantitative analysis in untargetted metabolomics since tandem mass spectrometry (MS/MS) is rarely used for compound identification. Therefore shortcomings in the data processing steps can easily introduce false positives due to misalignments and erroneous normalization adjustments in large sample studies. RESULTS: In this work, we developed a software package MetTailor featuring two novel data preprocessing steps to remedy drawbacks in the existing processing tools. First, we propose a novel dynamic block summarization (DBS) method for correcting misalignments from peak alignment algorithms, which alleviates missing data problem due to misalignments. For the purpose of verifying correct re-alignments, we propose to use the cross-sample consistency in isotopic intensity ratios as a quality metric. Second, we developed a flexible intensity normalization procedure that adjusts normalizing factors against the temporal variations in total ion chromatogram (TIC) along the chromatographic retention time (RT). We first evaluated the DBS algorithm using a curated metabolomics dataset, illustrating that the algorithm identifies misaligned peaks and correctly realigns them with good sensitivity. We next demonstrated the DBS algorithm and the RT-based normalization procedure in a large-scale dataset featuring >100 sera samples in primary Dengue infection study. Although the initial alignment was successful for the majority of peaks, the DBS algorithm still corrected ∼7000 misaligned peaks in this data and many recovered peaks showed consistent isotopic patterns with the peaks they were realigned to. In addition, the RT-based normalization algorithm efficiently removed visible local variations in TIC along the RT, without sacrificing the sensitivity of detecting differentially expressed metabolites. AVAILABILITY AND IMPLEMENTATION: The R package MetTailor is freely available at the SourceForge website http://mettailor.sourceforge.net/. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

UNLABELLED: We present LuciPHOr2, a site localization tool for generic post-translational modifications (PTMs) using tandem mass spectrometry data. As an extension of the original LuciPHOr (version 1) for phosphorylation site localization, the new software provides a site-level localization score for generic PTMs and associated false discovery rate called the false localization rate. We describe several novel features such as operating system independence and reduced computation time through multiple threading. We also discuss optimal parameters for different types of data and illustrate the new tool on a human skeletal muscle dataset for lysine-acetylation. AVAILABILITY AND IMPLEMENTATION: The software is freely available on the SourceForge website http://luciphor2.sourceforge.net. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Multiply correlated datasets have become increasingly common in genome-wide location analysis of regulatory proteins and epigenetic modifications. Their correlation can be directly incorporated into a statistical model to capture underlying biological interactions, but such modeling quickly becomes computationally intractable. RESULTS: We present sparsely correlated hidden Markov models (scHMM), a novel method for performing simultaneous hidden Markov model (HMM) inference for multiple genomic datasets. In scHMM, a single HMM is assumed for each series, but the transition probability in each series depends on not only its own hidden states but also the hidden states of other related series. For each series, scHMM uses penalized regression to select a subset of the other data series and estimate their effects on the odds of each transition in the given series. Following this, hidden states are inferred using a standard forward-backward algorithm, with the transition probabilities adjusted by the model at each position, which helps retain the order of computation close to fitting independent HMMs (iHMM). Hence, scHMM is a collection of inter-dependent non-homogeneous HMMs, capable of giving a close approximation to a fully multivariate HMM fit. A simulation study shows that scHMM achieves comparable sensitivity to the multivariate HMM fit at a much lower computational cost. The method was demonstrated in the joint analysis of 39 histone modifications, CTCF and RNA polymerase II in human CD4+ T cells. scHMM reported fewer high-confidence regions than iHMM in this dataset, but scHMM could recover previously characterized histone modifications in relevant genomic regions better than iHMM. In addition, the resulting combinatorial patterns from scHMM could be better mapped to the 51 states reported by the multivariate HMM method of Ernst and Kellis. AVAILABILITY: The scHMM package can be freely downloaded from http://sourceforge.net/p/schmm/ and is recommended for use in a linux environment.

MOTIVATION: Chromatin immunoprecipitation (ChIP) experiments followed by array hybridization, or ChIP-chip, is a powerful approach for identifying transcription factor binding sites (TFBS) and has been widely used. Recently, massively parallel sequencing coupled with ChIP experiments (ChIP-seq) has been increasingly used as an alternative to ChIP-chip, offering cost-effective genome-wide coverage and resolution up to a single base pair. For many well-studied TFs, both ChIP-seq and ChIP-chip experiments have been applied and their data are publicly available. Previous analyses have revealed substantial technology-specific binding signals despite strong correlation between the two sets of results. Therefore, it is of interest to see whether the two data sources can be combined to enhance the detection of TFBS. RESULTS: In this work, hierarchical hidden Markov model (HHMM) is proposed for combining data from ChIP-seq and ChIP-chip. In HHMM, inference results from individual HMMs in ChIP-seq and ChIP-chip experiments are summarized by a higher level HMM. Simulation studies show the advantage of HHMM when data from both technologies co-exist. Analysis of two well-studied TFs, NRSF and CCCTC-binding factor (CTCF), also suggests that HHMM yields improved TFBS identification in comparison to analyses using individual data sources or a simple merger of the two. AVAILABILITY: Source code for the software ChIPmeta is freely available for download at http://www.umich.edu/~hwchoi/HHMMsoftware.zip, implemented in C and supported on linux.

BACKGROUND: With the explosion in data generated using microarray technology by different investigators working on similar experiments, it is of interest to combine results across multiple studies. RESULTS: In this article, we describe a general probabilistic framework for combining high-throughput genomic data from several related microarray experiments using mixture models. A key feature of the model is the use of latent variables that represent quantities that can be combined across diverse platforms. We consider two methods for estimation of an index termed the probability of expression (POE). The first, reported in previous work by the authors, involves Markov Chain Monte Carlo (MCMC) techniques. The second method is a faster algorithm based on the expectation-maximization (EM) algorithm. The methods are illustrated with application to a meta-analysis of datasets for metastatic cancer. CONCLUSION: The statistical methods described in the paper are available as an R package, metaArray 1.8.1, which is at Bioconductor, whose URL is http://www.bioconductor.org/.