Recent advances in sequencing technologies have allowed us to capture various aspects of the genome at single-cell resolution. However, with exception a few co-assaying technologies, it is not possible simultaneously apply different assays on same single cell. In this scenario, computational integration multi-omic measurements crucial enable joint analyses. This task particularly challenging due lack sample-wise or feature-wise correspondences. We present alignment optimal transport (SCOT),...

Abstract Data integration of single-cell measurements is critical for understanding cell development and disease, but the lack correspondence between different types makes such efforts challenging. Several unsupervised algorithms can align heterogeneous in a shared space, enabling creation mappings single cells data domains. However, these require hyperparameter tuning high-quality alignments, which difficult an setting without information validation. We present Single-Cell alignment using...

Multiomic single-cell data allow us to perform integrated analysis understand genomic regulation of biological processes. However, most sequencing assays are performed on separately sampled cell populations, as applying them the same is challenging. Existing unsupervised alignment algorithms have been primarily benchmarked coassay experiments. Our investigation revealed that these methods do not well for noncoassay experiments when there disproportionate cell-type representation across...

In the United States, public has a constitutional right to access criminal trial proceedings. practice, it can be difficult or impossible for exercise this right. We present JUSTFAIR: Judicial System Transparency through Federal Archive Inferred Records, database of sentencing decisions made in federal district courts. have compiled data set from sources including States Sentencing Commission, Center, Public Access Court Electronic Records system, and Wikipedia. With nearly 600,000 records...

Although the availability of various sequencing technologies allows us to capture different genome properties at single-cell resolution, with exception a few co-assaying technologies, applying assays on same single cell is impossible. Single-cell alignment using optimal transport (SCOT) an unsupervised algorithm that addresses this limitation by align multiomics data. First, it preserves local geometry constructing k-nearest neighbor (k-NN) graph for each data set (or domain) intra-domain...

Abstract Integrated analysis of multi-omics data allows the study how different molecular views in genome interact to regulate cellular processes; however, with a few exceptions, applying multiple sequencing assays on same single cell is not possible. While recent unsupervised algorithms align single-cell multi-omic datasets, these methods have been primarily benchmarked co-assay experiments rather than more common taken from separately sampled populations. Therefore, most existing perform...

In the Unites States, public has a constitutional right to access criminal trial proceedings. practice, it can be difficult or impossible for exercise this right. We present JUSTFAIR: Judicial System Transparency through Federal Archive Inferred Records, database of sentencing decisions made in federal district courts. have compiled data set from sources including United States Sentencing Commission, Center, Public Access Court Electronic Records system, and Wikipedia. With nearly 600,000...

Substantial race-based disparities exist in federal criminal sentencing. We analyze 380,000 recent (2006-2019) sentences the JUSTFAIR database and show that these are large vary considerably across judges. Judges assign white defendants 13% shorter than Black defendants’ 19% Hispanic defendants’, on average, conditional case characteristics district. one standard deviation above average their estimated Black-white disparity give 39% conditionally longer vis-à-vis an of 13%. Hispanic-white...

Studying 380,000 criminal cases in federal district courts from 2006 to 2019, we replicate previous findings that aggregate, conditional racial disparities sentence lengths are large. We further show estimated sentencing vary considerably across judges. Results suggest judges assign white defendants sentences conditionally 13% shorter than Black defendants’ and 19% Hispanic defendants’, on average. A judge who is one standard deviation above average terms of Black-white disparity gives 39%...

To study discrimination in automated decision-making systems, scholars have proposed several definitions of fairness, each expressing a different fair ideal. These require practitioners to make complex decisions regarding which notion employ and are often difficult use practice since they binary judgement system is or unfair instead explaining the structure detected unfairness. We present an optimal transport-based approach fairness that offers interpretable quantifiable exploration bias its...

Equation-free modeling aims at extracting low-dimensional macroscopic dynamics from complex high-dimensional systems that govern the evolution of microscopic states. This algorithm relies on lifting and restriction operators map states to vice versa. Combined with simulations state, this can be used apply Newton solvers implicitly defined system or solve it more efficiently using direct numerical simulations. The key challenge is construction restrictions usually require a priori insight...