September 21, 2017, Thursday

From BIMIB

Jump to: navigation, search

NGSMilan20120717

NGS Milan Meeting

Martedi’ 17 Luglio 2012 14.30-18.30, aula seminari, U14, Dipartimento di Informatica, Sistemistica e Comunicazione dell’Universita’ degli Studi Milano Bicocca, Viale Sarca, 336 Milano


Programma

  • 14,30-14,40
NGS Milano News: what’s new; job board; presentation of the talks & of the speakers.
  • 14,40-15,00 - NGS Methods Talk I
Title: Aspetti algoritmici di base nell'analisi di dati NGS
Speaker: Raffaella Rizzi
Affiliation: Disco – Bicocca
Abstract:
Le nuove tecnologie di sequenziamento (Next-Generation Sequencing, NGS) producono grandi quantità di dati con caratteristiche completamente diverse rispetto ai dati disponibili in precedenza. In particolare la lunghezza dei read è decisamente minore e la distribuzione degli errori dipende fortemente dalla tecnologia utilizzata.
Questi aspetti, insieme alla dimensione dei dati trattati, porta i vecchi metodi disegnati per le sequenze di tipo tradizionale (Sanger) a fallire se applicati ai dati NGS, anche a causa dell'eccessivo tempo di calcolo e dell'elevato utilizzo di memoria. Di conseguenza nuovi approcci sono stati recentemente introdotti, ma il loro studio può essere considerato solo come preliminare, proprio a causa della recente introduzione.
Questo talk si propone di presentare e discutere alcuni aspetti fondazionali di natura computazionale e algoritmica legati alle tecnologie NGS.


  • 15,00-15,20 - NGS Discovery Talk I
Title: Characterization and validation of InDel sequence variants in 173 patient exomes
Speaker: Silvia Bonfiglio
Affiliation: HSR
Abstract:
We performed an extensive validation by standard dideoxy-sequencing of the common, rare and private indel variants identified within data generated by Illumina sequencing of 173 exomes from different diseases. While common variants present higher false positive rates, in line with variant validation issues presented by other groups, almost all private variants and a significant proportion of rare variants validated. These results suggest a potential role for private and rare Loss of Function (LoF) indels in disease genetics.


  • 15,20 -15,50 New NGS Technologies
Title: Le nuove frontiere Lifetech del sequenziamento massivo ad altissima densità: Wildfire e Proton per SOLiD ed Ion Torrent
Speaker: Raimo Tanzi
Affiliation: Lifetech
Abstract:
Wildfire è un metodo alternativo all'emulsion PCR per generare templati sulla piattaforma SOLiD 5500 basato su amplificazione isoterma in situ. in due ore genera clusters a densità tripla rispetto all'attuale mantenendo la flesibilità delle lanes indipendenti. I chip Ion torrent di seconda generazione incrementano la densità di un fattore 10 e 100 rendendo la tecnologia scalabile da 10 megabasi a quasi 100 Gigabasi. Con il Sistema Ampliseq, Ion Torrent offre una soluzione integrata, rapida e semplice per l'analisi di pannelli di geni umani.


  • 15,50-16,10 - NGS Methods Talk II
Title: Analysis workflow for the identification of allelic variants associated with rare disorders using WGS approach
Speaker: Vincenza Maselli
Affiliation: HSR
Abstract:
Whole Genome Sequencing at reasonable coverage has turned into a cost-effective technology to characterize the genetic framework of some diseases. We present an example of a case of WGS analysis in a case of a rare recessive disease.


  • 16,10-16,30 - NGS Discovery Talk II
Title: Ancient DNA study by Next Generation Sequencing
Speaker: Ermanno Rizzi
Affiliation: ITB-CNR
Abstract:
Next-generation sequencing, make it now possible the recovery of ancient DNA (aDNA) information from archaeological and paleontological remains, allowing us to go back in time and to study the genetic relationships between extinct organisms and their contemporary relatives. With the next-generation sequencing methodologies and strategies for genetic enrichment, DNA sequences can be retrieved even from samples (for example human remains) for which the technical pitfalls of classical methodologies required stringent criteria to guarantee the reliability of the results.


  • 16,30-17,00 - Coffee Break & informal discussions with the speakers


  • 17,00-17,20 - NGS Discovery Talk III
Title: Exploring the Human Sperm Head Transcriptome
Speaker: Alessandro Guffanti
Affiliation: Genomnia
Abstract:
Infertility affects one in six couples of childbearing age, and male problems are responsible for 40% of these cases. Infertility is becoming a public health issue as birth rates decline across Europe. Over the last 50 years, they have dropped to a level of 1.4 children per couple. The molecular details of the role of sperm head RNA in the process of fertilization have always been elusive, but there is growing evidence of an important role of the transcripts, including ncRNAs, shuttled by the sperm to the oocyte in the correct fertilization process. We present here the results of a comparative NGS exploration of human sperm head RNA samples from fertile and sterile individuals; these results highlight the complexity, heterogeneity and fragility of the first steps of human fertilization mechanisms as mediated by human germinal cells.


  • 17,20-17,40 - NGS Methods Talk III
Title: NGS data sharing, visualization, and analysis using Genopub/Integrated Genome Browser software
Speaker: Heiko Muller
Affiliation: IIT – SEMM
Abstract:
The predominant mode of visualizing NGS data is based on generating custom tracks in the UCSC genome browser. The UCSC genome browser uses a thin client such as Firefox or Safari to display the data that are assembled server-side into a clickable image using the html MAP tag. This mode of operation has obvious limitations. For example, data can only be displayed and not analyzed on the fly, zooming in and out, as well as navigating the genome requires roundtrips over the network, resulting in a discontinuous browsing experience. Tools like JBrowse have been devised that use Google Maps inspired pre-rendering of data on the server-side along with AJAX technology to allow for smoother genome browsing.
Another approach has been the development of dedicated genome browsing clients (“thick clients”) such as the Integrative Genome Viewer (IGV) or the Integrated Genome Browser (IGB). Both of these tools can retrieve annotations from Distributed Annotation System ([www.biodas.org DAS]) servers. This capability is most advanced in the IGB, which implements retrieval of large genomic data sets via the DAS/2 protocol. The DAS/2 protocol is distinguished from the classical DAS protocol by allowing custom (non-XML) based data formats. The reference implementation of the DAS/2 protocol is the Genopub genomics server ([http://bioserver.hci.utah.edu/BioInfo/index.php/Software:GenoPubInstall Genopub). We are using the Genopub/Integrated Genome Browser software packages to display and share genomic data. The recent introduction of a plugin interface in the Integrated Genome Browser and the collection of plugins that are already available herald the transformation of genome browsers from being mere visualization tools into fully fledged genomic data analysis suites. Advantages and limitations of the Genopub/Integrated Genome Browser software packages will be illustrated.


  • 17,40-18,00 - NGS Methods Talk IV
Title: Exploring PLS-DA approach in the analysis of RNA-seq experiments
Speaker: Jose Garcia Manteiga
Affiliation: HSR
Abstract:
Despite being widely used for identifying discriminant features in metabolomics experiments, partial least squares (projection latent structures) regression and discriminant analysis (PLS-DA) has not been exploited equally when applied to gene expression analysis. PLS-DA does not assume ordered categories and hence is more flexible to yield biologically meaningful results than other techniques. Moreover, albeit being a rather simple statistical technique, PLS is specially designed to withstand departures from normality and deal with multicollineality with robustness, often present in gene expression analyses. Although some attempts had been done to show PLS-DA performance with microarray data, nothing has been done in the case of digital gene expression data. We have compared PLS-DA with the established limma/edgeR procedure for extracting relevant features using two different case studies.


  • 18,00-18,30 Q&A Session
Q&A session for all the speakers