Subarna Sinha

Epigenetic and in vivo comparison of diverse MSC sources reveals an endochondral signature for human hematopoietic niche formation

In the last decade there has been a rapid expansion in clinical trials using mesenchymal stromal cells (MSCs) from a variety of tissues. However, despite similarities in morphology, immunophenotype, and differentiation behavior in vitro, MSCs sourced from distinct tissues do not necessarily have equivalent biological properties. We performed a genome-wide methylation, transcription, and in vivo evaluation of MSCs from human bone marrow (BM), white adipose tissue, umbilical cord, and skin cultured in humanized media. Surprisingly, only BM-derived MSCs spontaneously formed a BM cavity through a vascularized cartilage intermediate in vivo that was progressively replaced by hematopoietic tissue and bone. Only BM-derived MSCs exhibited a chondrogenic transcriptional program with hypomethylation and increased expression of RUNX3, RUNX2, BGLAP, MMP13, and ITGA10 consistent with a latent and primed skeletal developmental potential. The humanized MSC-derived microenvironment permitted homing and maintenance of long-term murine SLAM(+) hematopoietic stem cells (HSCs), as well as human CD34(+)/CD38(-)/CD90(+)/CD45RA(+) HSCs after cord blood transplantation. These studies underscore the profound differences in developmental potential between MSC sources independent of donor age, with implications for their clinical use. We also demonstrate a tractable human niche model for studying homing and engraftment of human hematopoietic cells in normal and neoplastic states.

Accurate process-hotspot detection using critical design rule extraction

In advanced fabrication technology, the sub-wavelength lithography gap causes unwanted layout distortions. Even if a layout passes design rule checking (DRC), it still might contain process hotspots, which are sensitive to the lithographic process. Hence, process-hotspot detection has become a crucial issue. In this paper, we propose an accurate process-hotspot detection framework. Unlike existing DRC-based works, we extract only critical design rules to express the topological features of hotspot patterns. We adopt a two-stage filtering process to locate all hotspots accurately and efficiently. Compared with state-of-the-art DRC-based works, our results show that our approach can reach 100% success rate with significant speedups.

Mutant WT1 is associated with DNA hypermethylation of PRC2 targets in AML and responds to EZH2 inhibition

Acute myeloid leukemia (AML) is associated with deregulation of DNA methylation; however, many cases do not bear mutations in known regulators of cytosine guanine dinucleotide (CpG) methylation. We found that mutations in WT1, IDH2, and CEBPA were strongly linked to DNA hypermethylation in AML using a novel integrative analysis of The Cancer Genome Atlas data based on Boolean implications, if-then rules that identify all individual CpG sites that are hypermethylated in the presence of a mutation. Introduction of mutant WT1 (WT1mut) into wild-type AML cells induced DNA hypermethylation, confirming mutant WT1 to be causally associated with DNA hypermethylation. Methylated genes in WT1mut primary patient samples were highly enriched for polycomb repressor complex 2 (PRC2) targets, implicating PRC2 dysregulation in WT1mut leukemogenesis. We found that PRC2 target genes were aberrantly repressed in WT1mut AML, and that expression of mutant WT1 in CD34(+) cord blood cells induced myeloid differentiation block. Treatment of WT1mut AML cells with short hairpin RNA or pharmacologic PRC2/enhancer of zeste homolog 2 (EZH2) inhibitors promoted myeloid differentiation, suggesting EZH2 inhibitors may be active in this AML subtype. Our results highlight a strong association between mutant WT1 and DNA hypermethylation in AML and demonstrate that Boolean implications can be used to decipher mutation-specific methylation patterns that may lead to therapeutic insights.

Improved tangent space based distance metric for accurate lithographic hotspot classification

A distance metric of patterns is crucial to hotspot cluster analysis and classification. In this paper, we propose an improved tangent space based metric for pattern matching based hotspot cluster analysis and classification. The proposed distance metric is an important extension of the well-developed tangent space method in computer vision. It can handle patterns containing multiple polygons, while the traditional tangent space method can only deal with patterns with a single polygon. It inherits most of the advantages of the traditional tangent space method, e.g., it is easy to compute and is tolerant with small variations or shifts of the shapes. Compared with the existing distance metric based on XOR of hotspot patterns, the improved tangent space based distance metric can achieve up to 37.5% accuracy improvement with at most 4.3× computational cost in the context of cluster analysis. The improved tangent space based distance metric is a more reliable and accurate metric for hotspot cluster analysis and classification. It is more suitable for industry applications.

Mining TCGA Data Using Boolean Implications

Boolean implications (if-then rules) provide a conceptually simple, uniform and highly scalable way to find associations between pairs of random variables. In this paper, we propose to use Boolean implications to find relationships between variables of different data types (mutation, copy number alteration, DNA methylation and gene expression) from the glioblastoma (GBM) and ovarian serous cystadenoma (OV) data sets from The Cancer Genome Atlas (TCGA). We find hundreds of thousands of Boolean implications from these data sets. A direct comparison of the relationships found by Boolean implications and those found by commonly used methods for mining associations show that existing methods would miss relationships found by Boolean implications. Furthermore, many relationships exposed by Boolean implications reflect important aspects of cancer biology. Examples of our findings include cis relationships between copy number alteration, DNA methylation and expression of genes, a new hierarchy of mutations and recurrent copy number alterations, loss-of-heterozygosity of well-known tumor suppressors, and the hypermethylation phenotype associated with IDH1 mutations in GBM. The Boolean implication results used in the paper can be accessed at http://crookneck.stanford.edu/microarray/TCGANetworks/.

Systematic discovery of mutation-specific synthetic lethals by mining pan-cancer human primary tumor data

Two genes are synthetically lethal (SL) when defects in both are lethal to a cell but a single defect is non-lethal. SL partners of cancer mutations are of great interest as pharmacological targets; however, identifying them by cell line-based methods is challenging. Here we develop MiSL (Mining Synthetic Lethals), an algorithm that mines pan-cancer human primary tumour data to identify mutation-specific SL partners for specific cancers. We apply MiSL to 12 different cancers and predict 145,891 SL partners for 3,120 mutations, including known mutation-specific SL partners. Comparisons with functional screens show that MiSL predictions are enriched for SLs in multiple cancers. We extensively validate a SL interaction identified by MiSL between the IDH1 mutation and ACACA in leukaemia using gene targeting and patient-derived xenografts. Furthermore, we apply MiSL to pinpoint genetic biomarkers for drug sensitivity. These results demonstrate that MiSL can accelerate precision oncology by identifying mutation-specific targets and biomarkers.

Fast and scalable parallel layout decomposition in double patterning lithography

For 32/22nm technology nodes and below, double patterning (DP) lithography has become the most promising interim solutions due to the delay in the deployment of next generation lithography (e.g., EUV). DP requires the partitioning of the layout patterns into two different masks, a procedure called layout decomposition. Layout decomposition is a key computational step that is necessary for double patterning technology. Existing works on layout decomposition are all single-threaded, which is not scalable in runtime and/or memory for large industrial layouts. This paper presents the first window-based parallel layout decomposition methods for improving both runtime and memory consumption. Experimental results are promising and show the presented parallel layout decomposition methods obtain upto 21x speedup in runtime and upto 7.5xreduction in peak memory consumption with acceptable solution quality.

Improved Tangent Space-Based Distance Metric for Lithographic Hotspot Classification

A distance metric of patterns is crucial to hotspot cluster analysis and classification. In this paper, we propose an improved tangent space (ITS)-based distance metric for hotspot cluster analysis and classification. The proposed distance metric is an important extension of the well-developed tangent space method in computer vision. It can handle patterns containing multiple polygons, while the traditional tangent space method can only deal with patterns with a single polygon. It inherits most of the advantages of the traditional tangent space method, e.g., it is easy to compute and is tolerant with small variations or shifts of the shapes. The ITS-based distance metric is a more reliable and accurate metric for hotspot cluster analysis and classification. We also propose a hierarchical density-based clustering method for hotspot clustering. It is more suitable for arbitrary shaped clusters.

Abstract A27: Systematic discovery of mutation-specific synthetic lethals by mining pan-cancer primary tumor data

Synthetic lethality, in which a single gene defect leads to dependency on a second gene that is otherwise not essential, is an attractive paradigm to identify targeted therapies for somatic mutations. Current methods to detect synthetic lethal (SL) partners for somatic mutations use large-scale shRNA screens in cell lines, combine shRNA data with tumor genomic data or use human orthologs of yeast SL interactions. These approaches are limited as they rely on cell line or yeast data, which are not representative of primary tumors. We have developed MiSL, a novel computational algorithm that utilizes large pan-cancer patient datasets (mutation, copy number and gene expression) to identify SL partners for specific mutations in specific cancer types. The underlying assumption of our approach is that, across multiple cancers, SL partners of a mutation will be amplified more frequently or deleted less frequently, with concordant changes in expression, in primary tumor samples harboring the mutation. Application of MiSL produced candidate SL partners for 30-80% of recurrent mutations in 12 cancers. Importantly, MiSL identified candidate SL partners for mutations (mut) in genes such as IDH1 that are not well-represented in existing cell lines. This is a distinct advantage over recent computational methods that combine shRNA data along with genomic data to make their predictions. Since MiSL uses only genomic and gene expression data, it allows assessment of a wide range of primary human tumors and mutations found in large primary tumor data sets such as TCGA. We validated MiSL using existing data and large-scale shRNA experiments we performed in doxycycline-inducible expression systems. We found that IDH1mut MiSL candidates in acute myeloid leukemia (AML) were enriched (p=0.004) for essential genes specific to IDH1mut but not IDH1 wildtype cells determined by a DECIPHER shRNA screen covering 9,965 human genes performed in doxycycline-inducible IDH1 (R132) THP-1 cells. Importantly, 1 out of 5 MiSL candidates was a SL partner of IDH1mut in AML cells as per the shRNA screen, indicating MiSL9s strong predictive power. Also, for multiple mutations in colorectal cancer, MiSL candidates were enriched (p Next, we used MiSL to identify novel and druggable SL partners in (i) AML and (ii) breast cancer. MiSL predicted a novel SL interaction in AML between IDH1mut and ACACA, the rate-limiting enzyme of fatty acid synthesis. Consistent with our prediction, pharmacologic or genetic blockade of ACACA prevented cell proliferation in the presence of IDH1mut, but not with IDH1 wildtype, in AML cell lines. Furthermore, when transduced with lentivirus encoding RFP-marked shRNA to ACACA, primary IDH1mut AML cells exhibited markedly reduced engraftment of RFP-positive human CD45+CD33+ leukemic cells compared to scrambled non-targeting shRNA (p In summary, MiSL is a general computational solution that finds novel SL interactions. Specifically, IDH1mut-ACACA is the first in vivo validated synthetic lethal in human tumor cells discovered purely by computational analysis of tumor genomic data. MiSL can greatly accelerate identification of pharmacologic targets associated with specific somatic mutations in specific tumor types for all kinds of mutations, thereby making it directly translatable to clinical applications. MiSL can also pinpoint predictive genetic biomarkers that can identify/extend indications for targeted therapies. Citation Format: Subarna Sinha, Daniel Thomas, Steven Chan, Yang Gao, Diede Brunen, Damoun Torabi, Andreas Reinisch, Rene Bernards, Ravindra Majeti, David L. Dill. Systematic discovery of mutation-specific synthetic lethals by mining pan-cancer primary tumor data. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Targeting the Vulnerabilities of Cancer; May 16-19, 2016; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(1_Suppl):Abstract nr A27.

Deciphering cancer biology using boolean methods

Boolean implications (if-then rules) provide a conceptually simple, uniform and highly scalable way to find associations between pairs of random variables. In this paper, we describe how Boolean implications can be derived from large, heterogeneous cancer data sets. We demonstrate two applications of Boolean implications to discover new actionable insights in cancer biology.