Data Analysis

Implementing a “Siamese” Neural Network with Mariana 1.0

Mariana was previously introduced in this blog by Geneviève in her May post Machine learning in life science. The Mariana codebase is currently standing on github at the third release candidate before the launch of the stable 1.0 release. This new version incorporates a large refactorization effort as well as many new features (a complete list of the changes found in the 1.0 version can be found in the changelog). I am taking this opportunity to present here a small tutorial on extending the [...]

By | 2016-11-25T09:29:10+00:00 November 7, 2016|Categories: Computer science, Data Analysis, Machine learning, Python|0 Comments

Bootstraps and Confidence Intervals

When analyzing data, you might want or need to fit a specific curve to a particular dataset. This type of analysis can result in instructive outputs regarding the relationship between two (or more...) quantifiable parameters. The main object of this post is not how to implement such fitting, but rather how to display the goodness of such a fit i.e. how to calculate a confidence interval around a fitted curve. That being said, I will show how to do curve fitting in [...]

By | 2016-11-08T09:30:03+00:00 September 29, 2016|Categories: Data Analysis, Data Visualization, R|1 Comment

Fastest method to compute an AUC

Context: AUC is an acronym for "Area Under the (ROC) Curve". If you are not familiar with the ROC curve and AUC, I suggest reading this blog post before to continuing further. For several projects, I needed to compute a large number of AUC. It started with 25,000, increased to 230,000 and now I need to compute 1,500,000 AUC. With so many AUC, the time to compute each one becomes critical. On the web, I don't find much information about this specific [...]

By | 2016-11-08T09:30:03+00:00 August 18, 2016|Categories: Data Analysis, Performance, Python, R, Statistics|0 Comments

SciPy and Logistic Regressions

Given a set of data points, we often want to see if there exists a satisfying relationship between them. Linear regressions can easily be visualized with Seaborn, a Python library that is meant for exploration and visualization rather than statistical analysis. As for logistic regressions, SciPy is a good tool when one does not have his or her own analysis script. Let's look at the optimize package                        from scipy.optimize import [...]

By | 2016-11-08T09:30:04+00:00 June 9, 2016|Categories: Bioinformatics, Data Analysis, Data Visualization, Python|0 Comments

Standard deviation on a correlation scatter plot

I was recently asked by a colleague to provide visualization of differential gene expression computed using RPKM values (two samples, no replicates) and highlight genes that were outside the distribution by 2 standard deviations or more. As a first draft, I quickly obliged by calculating the fold change distribution, computing standard deviation and drawing lines on either side of the diagonal to obtain: This turns out to be equivalent to computing the standard deviation of the residual of a linear [...]

By | 2016-11-08T09:30:06+00:00 April 5, 2016|Categories: Data Analysis, Data Visualization, R, Statistics|0 Comments