-
Generative adversarial networks (GAN)
Read more ...: Generative adversarial networks (GAN)GANs (Generative Adversarial Networks) are based on making two neural networks compete with each other. The first network, the generator, creates synthetic data from random noise (e.g. a sample normal…
-
RNA-Seq Analysis in WSL – Part 3 : Differential expression analysis using Ballgown
Read more ...: RNA-Seq Analysis in WSL – Part 3 : Differential expression analysis using BallgownOnce we create the transcript abundance data using Stringtie, we can now perform the differential expression analysis. With Strigtie, we had created the estimated transcript abundance data which is compatible…
-
RNA-Seq Analysis in WSL – Part 2 : Raw sequence reads to transcript abundance
Read more ...: RNA-Seq Analysis in WSL – Part 2 : Raw sequence reads to transcript abundanceIn this post we will see how to use HISAT2 and Stringtie to align raw RNA-Seq reads to annotated genome and estimate transcript abundances.
-
RNA-Seq Analysis in WSL – Part 1 : Installation of tools
Read more ...: RNA-Seq Analysis in WSL – Part 1 : Installation of toolsIn this series of posts, we will see how we can perform RNA-Seq analysis in Windows operating system using linux tools in WSL. Also, the part of analysis which require…
-
Find common genes from two conditions with their expression values using pandas merge method
Read more ...: Find common genes from two conditions with their expression values using pandas merge methodLet’s say we have two pandas dataframes. Each of them contain two columns, the gene name and their expression value. There may be other columns with more details of each…
-
OneHot encoding of columns in a pandas dataframe using get_dummies.
Read more ...: OneHot encoding of columns in a pandas dataframe using get_dummies.The simplest way I found for performing OneHot encoding is using pandas.get_dummies method. In just one line of code you can OneHot encode selected columns in a pandas dataframe. Also,…
-
Tensorflow: solution to memory exhaustion due to repeated loading of models
Read more ...: Tensorflow: solution to memory exhaustion due to repeated loading of modelsI had a few deep learning models saved. From a test data, I wanted to make subsets in specific way and evaluate them. Each subset was a combination of few…
-
Tensorflow: training loop from scratch
Read more ...: Tensorflow: training loop from scratchThe deep learning model has model weights which start from a random state wen the model object is created. As we expose the model with the training data, the weights…
-
How to plot metabolite concentrations in R?
Read more ...: How to plot metabolite concentrations in R?The data source Let’s say we have the final concentrations of following metabolites from three strains of bacteria: Each strain was grown in triplicates, so we have three values of…
-
How to save python objects using joblib
Read more ...: How to save python objects using joblibOften times you would want to save python objects for later use. For example, a dataset you constructed which could be used for several projects, or a transformer object with…
-
Histograms are the best way to visualize distribution of data points
Read more ...: Histograms are the best way to visualize distribution of data pointsData distribution plots help visualize how quantitative data points are spread over the range of their values. Distribution of quantitative data can be shown in various ways such as box-plots,…
-
Change elements of an array based on a condition using np.where
Read more ...: Change elements of an array based on a condition using np.whereLet’s say we want to convert multiple categorical variables into binary variables by selecting one category as “0” and the rest as “1”. Or we want to change the values…
