Sunday, October 28, 2012

Discovery of multi-dimensional modules by integrative analysis of cancer genomic data


http://nar.oxfordjournals.org/content/40/19/9379.abstract?etoc

Recent technology has made it possible to simultaneously perform multi-platform genomic profiling (e.g. DNA methylation (DM) and gene expression (GE)) of biological samples, resulting in so-called ‘multi-dimensional genomic data’. Such data provide unique opportunities to study the coordination between regulatory mechanisms on multiple levels. However, integrative analysis of multi-dimensional genomics data for the discovery of combinatorial patterns is currently lacking. Here, we adopt a joint matrix factorization technique to address this challenge. This method projects multiple types of genomic data onto a common coordinate system, in which heterogeneous variables weighted highly in the same projected direction form a multi-dimensional module (md-module). Genomic variables in such modules are characterized by significant correlations and likely functional associations. We applied this method to the DM, GE, and microRNA expression data of 385 ovarian cancer samples from the The Cancer Genome Atlas project. These md-modules revealed perturbed pathways that would have been overlooked with only a single type of data, uncovered associations between different layers of cellular activities and allowed the identification of clinically distinct patient subgroups. Our study provides an useful protocol for uncovering hidden patterns and their biological implications in multi-dimensional ‘omic’ data.                   

Neurobiology of resilience

http://www.nature.com/neuro/journal/v15/n11/full/nn.3234.html?WT.ec_id=NEURO-201211


Scott J Russo, James W Murrough, Ming-Hu Han, Dennis S Charney & Eric J Nestler

Nature Neuroscience 15, 1475–1484 (2012) doi:10.1038/nn.3234



Humans exhibit a remarkable degree of resilience in the face of extreme stress, with most resisting the development of neuropsychiatric disorders. Over the past 5 years, there has been increasing interest in the active, adaptive coping mechanisms of resilience; however, in humans, most published work focuses on correlative neuroendocrine markers that are associated with a resilient phenotype. In this review, we highlight a growing literature in rodents that is starting to complement the human work by identifying the active behavioral, neural, molecular and hormonal basis of resilience. The therapeutic implications of these findings are important and can pave the way for an innovative approach to drug development for a range of stress-related syndromes.

Saturday, October 27, 2012

Web programming

JSON Google API http://code.google.com/apis/gdata/json.html
Twitter Bootstrap front end http://twitter.github.com/bootstrap
Google App Engine

Sunday, October 21, 2012

Thursday, October 18, 2012

Brain Art

http://www.freilgen.de/FI-activities/FI-activ-albert.html

latexdiff - diff latex files

http://www.pppl.gov/~hammett/comp/tex/latexdiff.html

http://packages.ubuntu.com/lucid/latexdiff

$ sudo apt-get install latexdiff

~/bin/latexdiff old.tex new.tex > diff.tex
 
Git commands
 
# get version from 10 commits ago 
git show HEAD~10:chapter1_de.tex > chapter1_de_v1.tex
git diff HEAD HEAD@{10} chapter1_de.tex > diff-head-head10.txt 

How to make the most of transferable skills

Career toolkit: Ask the Expert

http://www.nature.com/naturejobs/science/career_toolkit/ask_the_expert?WT.ec_id=NATUREjobs-20121018

I am looking for an academic position and am facing an unusual situation: I am waiting for an answer from my first-choice employer, but have received an offer from a less-attractive research centre.


http://blogs.nature.com/naturejobs/2012/10/08/how-to-make-the-most-of-transferable-skills?WT.ec_id=NATUREjobs-20121018

Wednesday, October 17, 2012

Collaborations: The rise of research networks

http://www.nature.com/nature/journal/v490/n7420/full/490335a.html?WT.ec_id=NATURE-20121018

Nature
 
490,
 
335–336
 
(18 October 2012)
 
doi:10.1038/490335a
Published online
 

New collaboration patterns are changing the global balance of science. Established superpowers need to keep up or be left behind, says Jonathan Adams.

Tuesday, October 16, 2012

Chihaya Furu ちはやふる

http://en.wikipedia.org/wiki/Chihayafuru

ちはやふる

Chihayafuru (ちはやふる?) is a manga series written and illustrated by Yuki Suetsugu, serialised in Be Love and published by Kodansha. It is about a school girl, Chihaya Ayase, who is inspired by a new classmate to take up Hyakunin Isshu karuta competitively. It has been adapted into an anime television series, which aired on Nippon Television and Crunchyroll from 5 October 2011 to 28 March 2012. A second season will being airing in January 2013.

http://www.anime-mp3.com/2011/10/chihayafuru.html

Opening Theme:

YOUTHFUL by 99RadioService


Ending Theme:

"Soshite Ima" by Asami Seto

Sunday, October 14, 2012

Pediatric Imaging, Neurocognition, and Genetics - PING for short

http://ping.chd.ucsd.edu/

The overarching goal of the project is to create a large MRI and genetics data resource to be shared openly with the scientific community. The data resource will also include information about the developing mental and emotional functions of the children. Investigators on the project are studying 1400 children between the ages of 3 and 20 years so that links between genetic variation and developing patterns of brain connectivity can be examined. These data are critical for our understanding of emerging personality and mental abilities in children. One might say that PING is a study of the genetic and neural factors that contribute to individuality; understanding why we have different personalities and mental qualities is critically important for solving many problems that affect children, including mental disorders, addictions, academic problems, and learning disabilities.

Bioinformatics Genome Glossary

http://www.ncbi.nlm.nih.gov/projects/genome/glossary.shtml

Bioinformatics Insight

http://www.microbeworld.org/index.php?option=com_jlibrary&view=article&id=3728


Introduction to bioinformatics with Dr. Steve Jones, Head, Bioinformatics, Genome Sciences Centre, BC Cancer Agency.

GGobi - rggobi and ggobi

GGobi

GGobi is an open source visualization program for exploring high-dimensional data. It provides highly dynamic and interactive graphics such as tours, as well as familiar graphics such as the scatterplot, barchart and parallel coordinates plots. Plots are interactive and linked with brushing and identification.

http://www.ggobi.org/

> install.packages('rggobi')
> library('rggobi')
> g <- ggobi(mtcars)
> ggobi_display_save_picture(displays(g)[[1]], "test.png")

http://www.youtube.com/watch?v=OIkuQFQLMeo

Thursday, October 11, 2012

Five top tips to starting a successful business

http://www.linkedin.com/today/post/article/20121002115242-204068115-five-top-tips-to-starting-a-successful-business?trk=NUS_UNIU_PEOPLE_FOLLOW-megaphone


1. Listen more than you talk
2. Keep it simple
3. Take pride in your work
4. Have fun, success will follow
5. Rip it up and start again - learn from failure

Genetics - null mutation

http://www.wormbook.org/chapters/www_epistasis.2/epistasis.html

a null mutation, a loss-of-function mutation, no functional transcript, no function at all

or a gain-of-function mutation

http://en.wikipedia.org/wiki/Nonsense_mutation
Many organisms—including humans and lower species, such as yeast -- employ a nonsense-mediated mRNA decay (NMD) pathway, which degrades mRNAs containing nonsense mutations before they are translated into nonfunctional polypeptides.

(also called an mRNA-surveillance pathway)

http://www.urmc.rochester.edu/labs/Maquat-Lab/
http://www.urmc.rochester.edu/labs/Maquat-Lab/projects/nonsense-mediated_mrna_decay_nmd

http://www.youtube.com/watch?v=oBg6Wpab81M

http://www.answers.com/topic/mrna-surveillance
the process occurring in the cytosol of eukaryotic cells whereby mRNA that contains a premature stop codon is detected and submitted to selective degradation by nonsense-mediated mRNA decay. It involves assembly of a surveillance complex on the mRNA about 24 nt upstream of exon-exon junctions. If translation terminates because of an upstream stop codon, translation release factors bind to the surveillance complex and trigger mRNA degradation.

Wednesday, October 10, 2012

Mental health: Under a cloud

http://www.nature.com/nature/journal/v490/n7419/full/nj7419-299a.html?WT.ec_id=NATURE-2012-10-11

Depression is rife among graduate students and postdocs. Universities are working to get them the help they need.

She developed a crippling fear of presenting her research. “Doing a PhD is such a personal thing, one that you've invested so much time in, that any criticism can feel like a direct reflection of yourself,” says Lauren.

But she did something that many postgraduates do not: she got help. With counselling and medication, Lauren — a pseudonym that she uses on a blog detailing her experience (see go.nature.com/4ta9fo) — is entering the final year of her PhD. Hers is one of more than 50 stories highlighted on the website Students Against Depression, funded by the Charlie Waller Memorial Trust in Thatcham, UK. “The website aims to raise awareness that depression isn't a personal failing or weakness; it's a serious condition that requires treatment,” says psychologist Denise Meyer, the website's project manager.

“Getting help,” she says, “is a sign of strength, not weakness.”

Sparse PLS discriminant analysis: biologically relevant feature selection and graphical displays for multiclass problems

http://www.biomedcentral.com/1471-2105/12/253

Abstract

Background

Variable selection on high throughput biological data, such as gene expression or single nucleotide polymorphisms (SNPs), becomes inevitable to select relevant information and, therefore, to better characterize diseases or assess genetic structure. There are different ways to perform variable selection in large data sets. Statistical tests are commonly used to identify differentially expressed features for explanatory purposes, whereas Machine Learning wrapper approaches can be used for predictive purposes. In the case of multiple highly correlated variables, another option is to use multivariate exploratory approaches to give more insight into cell biology, biological pathways or complex traits.

Results

A simple extension of a sparse PLS exploratory approach is proposed to perform variable selection in a multiclass classification framework.

Conclusions

sPLS-DA has a classification performance similar to other wrapper or sparse discriminant analysis approaches on public microarray and SNP data sets. More importantly, sPLS-DA is clearly competitive in terms of computational efficiency and superior in terms of interpretability of the results via valuable graphical outputs. sPLS-DA is available in the R package mixOmics, which is dedicated to the analysis of large biological data sets.

Friday, October 5, 2012

WHY in science

Andy Torr told us that the key to engaging our audience is by explaining the "WHY" - WHY are you so passionate about your work and WHY does it matter to your audience? This simple rule is at the heart of science communication. More importantly, when speaking to potential employers or writing your biographies, conveying your passion will become much more powerful. Andy provided tools for avoiding jargon and distilled effective communication strategies into elegant models. The event reached a climax when we had the opportunity to interact with each other and integrate the strategies in our elevator pitch."

http://thesbn.ca/sbncriticalskillsworkshopseries2012

Random forest

http://www.nature.com/nrg/journal/v10/n6/box/nrg2579_BX2.html

Rather than using a single classification tree, substantial improvements in classification accuracy can result from growing an ensemble of trees and letting them 'vote' for the most popular outcome class, given a set of input variable values. Such ensemble approaches can be used to provide measures of variable importance, a feature that is of great interest in genetic studies and that is often lacking in machine-learning approaches. The most widely used ensemble tree approach is probably the random forests method75. A random forest is constructed by drawing with replacement several bootstrap samples of the same size (for example, the same number of cases and controls) from the original sample. An unpruned classification tree is grown for each bootstrap sample, but with the restriction that at each node, rather than considering all possible predictor variables, only a random subset of the possible predictor variables is considered.