Posts Tagged ‘ans’

Visualization of Statistical Power Analysis

Thursday, July 28th, 2016

Visualization of Power Analysis Useful sliders giving one a feel of the #statistics

Babies and cats as coauthors on a manuscript

Sunday, November 29th, 2015

Discussion about listing babies & cats as co-authors on scientific
manuscripts via @bornalibran #authorship

Software licenses and retraction

Wednesday, November 25th, 2015

Treefinder Retraction Note [interestingly, done editorially due to change in software-license terms] HT @bornalibran

Panorama of ancient metazoan macromolecular complexes : Nature : Nature Publishing Group

Saturday, September 26th, 2015

Panorama of ancient metazoan #macromolecular complexes Finding many more #complexes from integrating many co-elutions

IBM Research: Preserving Validity in Adaptive Data Analysis

Wednesday, September 23rd, 2015

Preserving Validity in Adaptive Data Analysis Using differential #privacy for correct #stats even w/ test-set reuse

“A common next step would be to use the least-squares linear regression to check whether a simple linear combination of the three strongly correlated foods can predict the grade. It turns out that a little combination goes a long way: we discover that a linear combination of the three selected foods can explain a significant fraction of variance in the grade (plotted below). The regression analysis also reports that the p-value of this result is 0.00009 meaning that the probability of this happening purely by chance is less than 1 in 10,000.

Recall that no relationship exists in the true data distribution, so this discovery is clearly false. This spurious effect is known to experts as Freedman’s paradox. It arises since the variables (foods) used in the regression were chosen using the data itself.

We found that challenges of adaptivity can be addressed using techniques developed for privacy-preserving data analysis. These techniques rely on the notion of differential privacy that guarantees that the data analysis is not too sensitive to the data of any single individual. We rigorously demonstrated that ensuring differential privacy of an analysis also guarantees that the findings will be statistically valid. We then also developed additional approaches to the problem based on a new way to measure how much information an analysis reveals about a dataset.

The Thresholdout Algorithm

Using our new approach we designed an algorithm, called Thresholdout, that allows an analyst to reuse the holdout set of data for validating a large number of results, even when those results are produced by an adaptive analysis.


seasonal effects on gene expression

Sunday, August 9th, 2015

Widespread [25% genes] seasonal…expression reveals [circ]annual differences in…immunity, relevant for vaccination

In addition to circadian rhythms, batch effects, now consider seasonal effects on gene expression

Illumina poster of -seq experiments

Sunday, August 9th, 2015

Poster of -Seq expts from @Illumina Nextgen update to @Roche’s famous biochem. #pathway chart

Originally Boehringer Mannheim chart of pathways

Single cell Genome + Transcriptome Sequencing

Tuesday, July 28th, 2015

G&T-seq: parallel…#singlecell genomes & transcriptomes by @CGATist
lab & others low cov. matched data on 130+ cells

G&T-seq: parallel sequencing of single-cell genomes and transcriptomes

Iain C Macaulay,
Wilfried Haerty,
Parveen Kumar,
Yang I Li,
Tim Xiaoming Hu,
Mabel J Teng,
Mubeen Goolam,
Nathalie Saurat,
Paul Coupland,
Lesley M Shirley,
Miriam Smith,
Niels Van der Aa,
Ruby Banerjee,
Peter D Ellis,
Michael A Quail,
Harold P Swerdlow,
Magdalena Zernicka-Goetz,
Frederick J Livesey,
Chris P Ponting
& Thierry Voet

Nature Methods 12, 519–522 (2015) doi:10.1038/nmeth.3370Received 18 November 2014 Accepted 27 March 2015 Published online 27 April 2015


Wednesday, June 10th, 2015


Found this review

Sunday, May 31st, 2015

Nice graph of seq. machine output v time