Biology subjects

Kasif, S.

Publications and source records attributed to Kasif, S..

4 recordsLinked to original sources

bioRxiv2021

Targeted in silico characterization of fusion transcripts in tumor and normal tissues via FusionInspector

MotivationGene fusions play a key role as driver oncogenes in tumors, and their reliable discovery and detection are important for cancer research, diagnostics, prognostics and guiding personalized therapy. While discovering gene fusions from genome sequencing can be laborious and costly, the resulting "fusion transcripts" can be recovered from RNA-seq data of tumor and normal samples. However, alleged and putative fusion transcripts can also arise from multiple sources other than chromosomal rearrangements, including cis- or trans-splicing events, experimental artifacts during RNA-seq or computational errors of transcriptome reconstruction methods. Understanding how to discern, interpret, categorize, and verify predicted fusion transcripts is essential for consideration in clinical settings and prioritization for further research. SummaryHere, we present FusionInspector for in silico characterization and interpretation of candidate fusion transcripts from RNA-seq and exploration of their sequence and expression characteristics. We applied FusionInspector to thousands of tumor and normal transcriptomes, and identified statistical and experimental features enriched among biologically impactful fusions. Through clustering and machine learning, we identified large collections of fusions potentially relevant to tumor and normal biological processes. We show that biologically relevant fusions are enriched for relatively high expression of the fusion transcript, imbalanced fusion allelic ratios, and canonical splicing patterns, and are deficient in sequence microhomologies detected between partner genes. We demonstrate that FusionInspector accurately validates fusion transcripts in silico, and helps identify and characterize numerous understudied fusions in tumor and normal tissues samples. FusionInspector is freely available as open source for screening, characterization, and visualization of candidate fusions via RNA-seq, and helps with transparent explanation and interpretation of machine learning predictions and their experimental sources. HighlightsO_LIFusionInspector software for supervised analysis of candidate fusion transcripts C_LIO_LIClustering of recurrent fusion transcripts resolves biologically relevant fusions C_LIO_LIIdentification of distinguishing characteristics of known and novel fusion transcripts in tumor and normal tissues C_LI

bioinformatics↗

bioRxiv2019

Novel mediation analysis of human plasma proteome and metabolome reveals mediators of improved glycemia after gastric bypass surgery

Molecular mechanisms by which Roux-en-Y gastric bypass (RYGB) improves glycemic control and metabolism in type 2 diabetes (T2D) remain incompletely understood. In the SLIMM-T2D trial, participants with T2D were randomized to RYGB or nonsurgical management and their fasting plasma proteome and metabolome were analyzed for up to 3 years. To identify analytes that mediate improvement in outcomes, we developed a high-throughput mediation analysis method (Hitman), which is significantly more powerful than existing methods. Top-ranking analyte mediators of glycemia improvement were growth hormone receptor and prolylhydroxyproline, which were more significant than any clinical mediator, including BMI. Beta-alanine and Histidine Metabolism (both including CNDP1) were top differentially regulated pathways, and Valine, Leucine and Isoleucine Degradation was also a top differentially-regulated pathway and a top mediator of improvement in insulin resistance. The identified analytes may serve as novel targets for T2D therapy. More broadly, Hitman can identify analyte mediators of outcomes in randomized trials for which high-throughput data are available.

systems biology↗

bioRxiv2017

Unexpected Properties of Short Genomic Tandem Repeats

Length polymorphisms in genomic short tandem repeats have been implicated in a variety of diseases, most notably human neurodegenerative disorders. Expansions of tandem repeats are also associated with genomic instability in cancer. Our previous study of length-3 tandem repeats uncovered a surprising pattern in the length distribution of certain such repeats in the non-coding regions of the human reference genome: a bias towards repeats of length 3n - 1, (n > 3). That is, the observed frequency of repeats of this length in the human genome is higher than expected by chance based on the frequency of shorter repeats.\n\nWe have hypothesized that this pattern may be a general property of genomic DNA. If true, this could have implications with regard to the dynamics of repeat expansion generally. To test this hypothesis, we have analyzed the genomic sequences of a broad range of eukaryotic organisms as well as several complete human genomes and obtained a number of thought provoking results. We establish that this unexpected elevation in frequency of 3n - 1 long repeats is statistically significant. We also expanded this analysis to different classes of genomic regions and tandem repeats of length four and five. The specific pattern was found in 13 of the 20 organisms analyzed, including all chordate and insect genomes tested. The bias pattern, however, was not confined to a single branch of the evolutionary tree. For some genomes, such as Drosophila melanogaster, the repeat bias surprisingly was also identified in exons. The pattern is present in both small and large genomes. A similar pattern was also found in tetranucleotide and pentanucleotide repeats in the human genome. Another surprising property was identified for the flanking GC content for triplet repeats of length 3n. These findings indicate a puzzling new genomic phenomenon with possible evolutionary and disease-related implications.

bioinformatics↗

bioRxiv2017

Not All Experimental Questions Are Created Equal: Accelerating Biological Data to Knowledge Transformation (BD2K) via Science Informatics, Active Learning and Artificial Intelligence

Pablo Picasso, when first told about computers, famously quipped \"Computers are useless. They can only give you answers.\" Indeed, the majority of effort in the first half-century of computational research has focused on methods for producing answers. Incredible progress has been achieved in computational modeling, simulation and optimization, across domains as diverse as astrophysics, climate studies, biomedicine, architecture, and chess. However, the use of computers to pose new questions, or prioritize existing ones, has thus far been quite limited.\n\nPicassos comment highlights the point that good questions can sometimes be more elusive than good answers. The history of science offers numerous examples of the impact of good questions. Paul Erd[o]s, the wandering monk of mathematical graph theory, offered small prizes for anyone who could prove conjectures he identified as important (1). The prizes varied in cash amounts based on the perceived complexity of the problem posed by Erd[o]s.\n\nPosing technical questions and allocating resources to answer them has taken on a new guise in the Internet age. The X-Prize foundation (http://www.xprize.org/) offers multi-million dollar bounties for grand technological goals, including goals for sequencing genomes or space exploration. Several companies provide portals where customers can place cash bounties on educational, scientific or technological challenges, while potential problem solvers can compete to produce the best solutions for these problems. Amazons Turk site (https://www.mturk.com/mturk/welcome) links people requesting performance of intellectual tasks to people willing to work on them for a fee. Such crowd-sourcing systems create markets of questions and answers, and can help allocate resources and capabilities efficiently.\n\nThis paradigm suggests a number of interesting questions for scientific research. In a resource limited environment, can funds and research capacity be allocated more efficiently? Can knowledge demand provide an alternative or complementary mechanism to traditional investigator-initiated research grants?\n\nThe fathers of Artificial Intelligence (AI) and Herbert Simon in particular envisioned the application of AI to Scientific Discovery in different forms and styles (focusing on physics). We follow on these early dreams and describe a novel approach aimed at remodeling of the biomedical research infrastructure and catalyze gene function determination. We aim to start a bold discussion of new ideas aimed towards increasing the efficiency of the allocation of research capacities, reproducibility, provenance tracking, removing redundancy and catalyzing knowledge gain with each experiment. In particular, we describe a tractable computational framework and infrastructure that can help researchers assess the potential information gain of millions of experiments before conducting them. The utility of experiments in this case is modeled as the predictive knowledge (formalized as information) to be gained as a result of performing the experiment. The experimentalist would then be empowered to select experiments that maximized information gain if they wished, recognizing that there are frequently other considerations, such as a specific technological or medical utility, that might over-ride the priority of maximizing information gain. The conceptual approach we develop is general, and here we apply it to the study of gene function.

bioinformatics↗