Ron Weiss

Bio: Ron Weiss is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Synthetic biology & Speech synthesis. The author has an hindex of 82, co-authored 292 publications receiving 89189 citations. Previous affiliations of Ron Weiss include French Institute for Research in Computer Science and Automation & Google.

Methods of evaluating gene expression levels

Abstract: Described herein are methods of evaluating the expression levels of DNA parts encoding proteins in test circuits. In particular, the methods disclosed herein are useful to evaluate the expression of an output protein regulated by a regulatory protein-genetic element pair.

Affinity Weighted Embedding

Abstract: Supervised (linear) embedding models like Wsabie and PSI have proven successful at ranking, recommendation and annotation tasks. However, despite being scalable to large datasets they do not take full advantage of the extra data due to their linear nature, and typically underfit. We propose a new class of models which aim to provide improved performance while retaining many of the benefits of the existing class of embedding models. Our new approach works by iteratively learning a linear embedding model where the next iteration's features and labels are reweighted as a function of the previous iteration. We describe several variants of the family, and give some initial results.

Customized signaling with reconfigurable protein scaffolds.

Abstract: VOLUME 26 NUMBER 5 MAY 2008 NATURE BIOTECHNOLOGY dengue viruses. These candidate vaccines seem promising, but studies are still underway to determine whether they can confer a balanced immune response against all four dengue viruses and avoid immunological interference. For West Nile virus, the lead candidate vaccines—a DNA vaccine and a chimeric yellow fever 17D virus—also appear promising, but there is still no consensus on what constitutes a long-term protective immune response. Chang et al.5 build on earlier work by Kofler et al.9,10, who showed that the tick-borne encephalitis flavivirus still forms immunogenic virus particles even when much of the capsid gene sequence has been deleted. Using Kunjin, a subtype of West Nile virus found in Australia, Chang et al.5 have developed a ‘split-genome’ vaccine that generates two RNA species, one encoding the entire Kunjin virus genome except the capsid gene and the other encoding only the capsid gene. As both RNAs are encoded on the same DNA plasmid under the control of two cytomegalovirus promoters configured in a back-toback orientation, transfected cells transcribe and translate all the viral genes. The capsid protein acts as a helper to assemble virus particles containing the viral genomic RNA lacking the capsid gene. These so-called single-round infectious particles (SRIPs) then infect adjacent cells (Fig. 1), in contrast to DNA vaccines that produce viral antigens only in the cells initially infected. Because the viral genome transmitted to neighboring cells does not encode capsid protein, no further viral replication can occur. Chang et al.5 compare the immunogenicity of SRIPs in mice to a live virus, a traditional DNA vaccine (encoding the viral genome, with the exception of functional capsid) and a DNA vaccine that produces virus-like particles composed of the pre-membrane and envelope proteins. SRIPs confer a superior antibodymediated immune response in mice and horses, as well as protective immunity in mice, at lower doses of DNA compared with the traditional DNA vaccine. CD8+ T-cell responses elicited by SRIPs in mice were also significantly greater than those produced by the virus-like particle vaccine, although smaller than those following immunization with live virus. Neutralizing antibodies are considered critical for achieving protective immunity, but it is clear that a vaccine must elicit both antibodyand cell-mediated immunity to ensure long-term protection. Although these results represent an important proof of concept of a technology that should in theory be applicable to any flavivirus, a couple of important points should be considered. First, comparison of DNA-based vaccine strategies is very difficult given the many variables involved (e.g., viral strain, viral gene(s) selected, different parental virus strains and codon optimization). Second, it remains to be seen whether the present findings translate to primates. Several candidate DNA vaccines have performed impressively in lower animals only to disappoint in clinical trials. Prospects for using a SRIP-based approach in veterinary vaccines, such as those against Japanese encephalitis and West Nile virus infections of horses, seem more promising in the short term, especially as killed vaccines do not induce long-term protective immunity and booster doses are required to maintain immunity. The major issues surrounding new candidate vaccines always concern efficacy and safety. With regard to efficacy, we know that current licensed flavivirus vaccines have neutralizing antibody as the correlate of protection and that only low levels of neutralizing antibodies are required for protective immunity. We do not know whether this will be true for vaccines against dengue and West Nile viruses—and even if it is, as most investigators believe, it is unclear what level of neutralizing antibodies will be required. This question is particularly complicated for a dengue vaccine, as the disease is caused by four genetically and serologically related viruses. For a tetravalent vaccine, higher levels of neutralizing antibodies might be needed to control four viruses simultaneously. Candidate vaccines, such as those involving SRIPs, may help achieve this goal, possibly through a prime-boost regimen, although it remains to be shown that SRIP-based vaccines are effective over the long term. In the 21st century, safety has become the paramount attribute of a vaccine, even more so then efficacy, as society will not accept any adverse events associated with a vaccine. In this study, Chang et al.5 have boosted efficacy using viral particles that have clear safety advantages over live attenuated vaccines.

Content-Based Methods for Knowledge Discovery in Music

Abstract: This chapter presents several computational approaches aimed at supporting knowledge discovery in music. Our work combines data mining, signal processing and data visualization techniques for the automatic analysis of digital music collections, with a focus on retrieving and understanding musical structure.

Cholesterol‐Amino‐Phosphate (CAP) Derived Lipid Nanoparticles for Delivery of Self‐Amplifying RNA and Restoration of Spermatogenesis in Infertile Mice

Abstract: Male infertility caused by genetic mutations is an important type of infertility. Currently, there is no reliable method in the clinic to address this medical need. The emergence of mRNA therapy provides a possible strategy for restoring mutant genes in the reproductive system. However, effective delivery of mRNA to spermatocytes remains a formidable challenge. Here a series of cholesterol‐amino‐phosphate (CAP) lipids are reported by integrating three bioactive moieties into a geometric structure, which is favorable for mRNA delivery. The results demonstrate that CAP‐derived lipid nanoparticles (CAP LNPs) can deliver RNA including traditional mRNA and self‐amplifying RNA (saRNA) encoding DNA Meiotic Recombinase 1 (Dmc1) protein in spermatocytes and treat male infertility caused by the Dmc1 gene mutation. Notably, the delivery efficiency of CAP LNPs is significantly higher than that of the MC3 and ALC‐0315 LNPs, which is consistent with the design of CAP molecules. More importantly, a single injection of CAP LNPs–saRNA can produce Dmc1 protein for an extended period, which restores the spermatogenesis in the Dmc1 gene knockout mouse model. Overall, this study proves the concept of LNPs for the delivery of mRNA to spermatocytes, which provides a unique method to probe male infertility caused by the genetic mutation.

疟原虫var基因转换速率变化导致抗原变异[英]／Paul H, Robert P, Christodoulou Z, et al//Proc Natl Acad Sci U S A

Abstract: 抗原变异可使得多种致病微生物易于逃避宿主免疫应答。表达在感染红细胞表面的恶性疟原虫红细胞表面蛋白1（PfPMP1）与感染红细胞、内皮细胞、树突状细胞以及胎盘的单个或多个受体作用，在黏附及免疫逃避中起关键的作用。每个单倍体基因组var基因家族编码约60种成员，通过启动转录不同的var基因变异体为抗原变异提供了分子基础。

XGBoost: A Scalable Tree Boosting System

Abstract: Tree boosting is a highly effective and widely used machine learning method. In this paper, we describe a scalable end-to-end tree boosting system called XGBoost, which is used widely by data scientists to achieve state-of-the-art results on many machine learning challenges. We propose a novel sparsity-aware algorithm for sparse data and weighted quantile sketch for approximate tree learning. More importantly, we provide insights on cache access patterns, data compression and sharding to build a scalable tree boosting system. By combining these insights, XGBoost scales beyond billions of examples using far fewer resources than existing systems.

The anatomy of a large-scale hypertextual Web search engine

Abstract: In this paper, we present Google, a prototype of a large-scale search engine which makes heavy use of the structure present in hypertext. Google is designed to crawl and index the Web efficiently and produce much more satisfying search results than existing systems. The prototype with a full text and hyperlink database of at least 24 million pages is available at http://google.stanford.edu/. To engineer a search engine is a challenging task. Search engines index tens to hundreds of millions of web pages involving a comparable number of distinct terms. They answer tens of millions of queries every day. Despite the importance of large-scale search engines on the web, very little academic research has been done on them. Furthermore, due to rapid advance in technology and web proliferation, creating a web search engine today is very different from three years ago. This paper provides an in-depth description of our large-scale web search engine -- the first such detailed public description we know of to date. Apart from the problems of scaling traditional search techniques to data of this magnitude, there are new technical challenges involved with using the additional information present in hypertext to produce better search results. This paper addresses this question of how to build a practical large-scale system which can exploit the additional information present in hypertext. Also we look at the problem of how to effectively deal with uncontrolled hypertext collections where anyone can publish anything they want.

The PageRank Citation Ranking : Bringing Order to the Web

Abstract: The importance of a Web page is an inherently subjective matter, which depends on the readers interests, knowledge and attitudes. But there is still much that can be said objectively about the relative importance of Web pages. This paper describes PageRank, a mathod for rating Web pages objectively and mechanically, effectively measuring the human interest and attention devoted to them. We compare PageRank to an idealized random Web surfer. We show how to efficiently compute PageRank for large numbers of pages. And, we show how to apply PageRank to search and to user navigation.

XGBoost: A Scalable Tree Boosting System

Abstract: Tree boosting is a highly effective and widely used machine learning method. In this paper, we describe a scalable end-to-end tree boosting system called XGBoost, which is used widely by data scientists to achieve state-of-the-art results on many machine learning challenges. We propose a novel sparsity-aware algorithm for sparse data and weighted quantile sketch for approximate tree learning. More importantly, we provide insights on cache access patterns, data compression and sharding to build a scalable tree boosting system. By combining these insights, XGBoost scales beyond billions of examples using far fewer resources than existing systems.