While the advance of deep learning has allowed to automate many tasks in bioimage analysis, quantifying key visual features of biological objects in an image, such as cells, organs, or tissues, is still a multi-step and laborious task. It requires image segmentation and definition of features of interest, which often might be image- and problem-specific. This approach requires image labeling and training of the segmentation method as well as manual feature design and implementation of dedicated procedures for their quantification. Here we propose a self-supervised learning (SSL) approach to encoding in microscopy images morphological features of molecular structures that play role in disease phenotype and patient clinical diagnosis. We encode super-resolution images of slit diaphragm (SD)—a specialized membrane between podocyte cells in kidney—in a high-dimensional embedding space in an unsupervised manner, without the need of image segmentation and feature quantification. We inspect the embedding space and demonstrate its relationship to the morphometric parameters of the SD estimated with a previously published method. The SSL-derived image representations additionally reflect the level of albuminuria—a key marker of advancement of kidney disease in a cohort of chronic kidney disease patients. Finally, the embeddings allow for distinguishing mouse model of kidney disease from the healthy subjects with a comparable accuracy to classification based on SD morphometric features. In a one step and label-free manner the SSL approach offers possibility to encode meaningful details in biomedical images and allow for their exploratory, unsupervised analysis as well as further fine-tuning for specialized supervised tasks.

Morphological alterations at the kidney filtration barrier increase intrinsic capillary wall permeability resulting in albuminuria. However, automated, quantitative assessment of these morphological changes has not been possible with electron or light microscopy. Here we present a deep learning-based approach for segmentation and quantitative analysis of foot processes in images acquired with confocal and super-resolution fluorescence microscopy. Our method, Automatic Morphological Analysis of Podocytes (AMAP), accurately segments podocyte foot processes and quantifies their morphology. AMAP applied to a set of kidney diseases in patient biopsies and a mouse model of focal segmental glomerulosclerosis allowed for accurate and comprehensive quantification of various morphometric features. With the use of AMAP, detailed morphology of podocyte foot process effacement was found to differ between categories of kidney pathologies, showed detailed variability between diverse patients with the same clinical diagnosis, and correlated with levels of proteinuria. AMAP could potentially complement other readouts such as various omics, standard histologic/electron microscopy and blood/urine assays for future personalized diagnosis and treatment of kidney disease. Thus, our novel finding could have implications to afford an understanding of early phases of kidney disease progression and may provide supplemental information in precision diagnostics.

Major genomic deletions in independent eukaryotic lineages have led to repeated ancestral loss of biosynthesis pathways for nine of the twenty canonical amino acids. While the evolutionary forces driving these polyphyletic deletion events are not well understood, the consequence is that extant metazoans are unable to produce nine essential amino acids (EAAs). Previous studies have highlighted that EAA biosynthesis tends to be more energetically costly, raising the possibility that these pathways were lost from organisms with access to abundant EAAs. It is unclear whether present-day metazoans can reaccept these pathways to resurrect biosynthetic capabilities that were lost long ago or whether evolution has rendered EAA pathways incompatible with metazoan metabolism. Here, we report progress on a large-scale synthetic genomics effort to reestablish EAA biosynthetic functionality in mammalian cells. We designed codon-optimized biosynthesis pathways based on genes mined from Escherichia coli. These pathways were de novo synthesized in 3 kilobase chunks, assembled in yeasto and genomically integrated into a Chinese hamster ovary (CHO) cell line. One synthetic pathway produced valine at a sufficient level for cell viability and proliferation. 13C-tracing verified de novo biosynthesis of valine and further revealed build-up of pathway intermediate 2,3-dihydroxy-3-isovalerate. Increasing the dosage of downstream ilvD boosted pathway performance and allowed for long-term propagation of second-generation cells in valine-free medium at 3.2 days per doubling. This work demonstrates that mammalian metabolism is amenable to restoration of ancient core pathways, paving a path for genome-scale efforts to synthetically restore metabolic functions to the metazoan lineage.

Motivation The synthesis of large DNA assemblies has applications in biotechnology, and can help us better understand genome biology. These large DNA assemblies are often constructed from many smaller DNA segments, and it is critical to assess that they are correctly assembled. One low cost and rapid method to ensure that the connection between each segment is correct is to use PCR with primer pairs that span assembly junctions. However, the design of PCR primers for large assemblies consisting of multiple segments, and therefore containing multiple assembly junctions, is a challenging process. Rule-based automation of the process often results in finding primers that satisfy general criteria, but are not necessarily the best fit for every particular junction. Results We have developed MenDEL – a web-based DNA design application, that provides a primer pair computation tool for multiple assembly junctions in such a way that for each junction we automatically pick the optimal pair of primers based on user specified constraints. Availability and Implementation The MenDEL application is available at https://mendel-isg.nyumc.org to registered users, and the code base for computing junction primers is available at https://github.com/MendelProject/PrimerOptimization .

Precise Hox gene expression is crucial for embryonic patterning. Intra- Hox transcription factor binding and distal enhancer elements have emerged as the major regulatory modules controlling Hox gene expression. However, quantifying their relative contributions has remained elusive. Here, we introduce “synthetic regulatory reconstitution,” a conceptual framework for studying gene regulation, and apply it to the HoxA cluster. We synthesized and delivered variant rat HoxA clusters (130 to 170 kilobases) to an ectopic location in the mouse genome. We found that a minimal HoxA cluster recapitulated correct patterns of chromatin remodeling and transcription in response to patterning signals, whereas the addition of distal enhancers was needed for full transcriptional output. Synthetic regulatory reconstitution could provide a generalizable strategy for deciphering the regulatory logic of gene expression in complex genomes.

Motivation Synthetic genomics as a field seeks to synthesize large regions of genomes from the ground up. Such large-scale projects, especially in complex genomes can rely on pre-existing BAC (Bacterial Artificial Chromosome) libraries as starting material to reduce cost. However, choosing BACs that cover long DNA regions, especially those that require many BACs, is a manual, idiosyncratic, time consuming, and error prone process. Automating this work would make the assembly of large DNA constructs more efficient. Results We have developed MenDEL – a web-based DNA design application, that provides efficient tools for finding BACs that cover long regions of interest and allow for sorting results based on multiple user defined criteria - total length, number of BACs, longest BAC. etc. In addition, it enables the user to find a combination of BACs from pre-existing libraries that cover a region of interest not found in any single BAC. Availability and Implementation MenDEL application is available to registered users at https://mendel-isg.nyumc.org , Java code used in the application to find BAC sets is available at https://github.com/MendelProject/BACFinder

The kidneys constantly filter enormous amounts of fluid, with almost complete retention of albumin and other macromolecules in the plasma. Diseases of podocytes at the kidney filtration barrier reduce the intrinsic permeability of the capillary wall resulting in albuminuria. However, direct quantitative assessment of the underlying morphological changes has previously not been possible. Here we developed a deep learning-based approach for segmentation of foot processes in images acquired with optical microscopy. Our method – Automatic Morphological Analysis of Podocytes (AMAP) – accurately segments foot processes and robustly quantifies their morphology. It also robustly determined morphometric parameters, at a Pearson correlation of r > 0.71 with a previously published semi-automated approach, across a large set of mouse tissue samples. The artificial intelligence algorithm wasWe applied the analysis to a set of human kidney disease conditions allowing comprehensive quantification of various underlying morphometric parameters. These results confirmed that when podocytes are injured, they take on a more simplified architecture and the slit diaphragm length is significantly shortened, resulting in a reduction in the filtration slit area and a loss of the buttress force of podocytes which increases the permeability of the glomerular basement membrane to albumin.

AbstractPreciseHoxgene expression is crucial for embryonic patterning. Intra-Hoxtranscription factor binding and distal enhancer elements have emerged as the major regulatory modes controllingHoxgene expression. However, quantifying their relative contributions has remained elusive. Here, we introduce ‘synthetic regulatory reconstitution’, a novel conceptual framework for studying gene regulation and apply it to theHoxAcluster. We synthesized and delivered variant ratHoxAclusters (130-170 kilobases each) to an ectopic location in the mouse genome. We find that aHoxAcluster lacking distal enhancers recapitulates correct patterns of chromatin remodeling and transcription in response to patterning signals, while distal enhancers are required for full transcriptional output. Synthetic regulatory reconstitution is a generalizable strategy to decipher the regulatory logic of gene expression in complex genomes.One-Sentence SummaryReconstitution of gene regulation using large DNA constructs unravels the regulatory logic of a developmental gene locus.

Age-related Macular Degeneration (AMD) is a leading cause of blindness in the developed world, especially in aging populations, and is therefore an important target for new therapeutic development. Recently, there have been several studies demonstrating strong associations between AMD and sites of heritable genetic variation at multiple loci, including a highly significant association at 10q26. The 10q26 risk region contains two genes, HTRA1 and ARMS2, both of which have been separately implicated as causative for the disease, as well as dozens of sites of non-coding variation. To date, no studies have successfully pinpointed which of these variant sites are functional in AMD, nor definitively identified which genes in the region are targets of such regulatory variation. In order to efficiently decipher which sites are functional in AMD phenotypes, we describe a general framework for combinatorial assembly of large ‘synthetic haplotypes’ along with delivery to relevant disease cell types for downstream functional analysis. We demonstrate the successful and highly efficient assembly of a first-draft 119kb wild-type ‘assemblon’ covering the HTRA1/ARMS2 risk region. We further propose the parallelized assembly of a library of combinatorial variant synthetic haplotypes covering the region, delivery and analysis of which will identify functional sites and their effects, leading to an improved understanding of AMD development. We anticipate that the methodology proposed here is highly generalizable towards the difficult problem of identifying truly functional variants from those discovered via GWAS or other genetic association studies.