FAQ - nicococo/scRNA GitHub Wiki

Common Errors

• Error while installing the package.

There is a possibility that installing Anaconda first will resolve the issue you have. Otherwise, look at the top of the website (or alternatively the readme file), there should be a tiny grey-green image that should say 'build passing'. This little image tells you that it is possible to install and test the package under a controlled environment using Travis-CI. If you want to know more, please have a look at the .travis.yml file which contains the command line arguments to install the scRNA package on a fresh computer.

• ERROR (scRNA-source.sh and scRNA-target.sh): Intel MKL FATAL ERROR: Cannot load libmkl_avx.so or libmkl_def.so.

If this error occurs and Anaconda open data science platform is used, then use the conda install mkl first.

• ERROR (scRNA-target.sh): Number of common genes must not be 0!

No gene id in the source data can be found in the target data. This can happen if source and target dataset use different descriptions of gene ids, e.g. Ensembl gene/transcript descriptors and Gene names.

• ERROR (scRNA-source.sh and scRNA-target.sh): Number of genes/transcripts in data and gene-ids must coincide.

While loading gene ids and corresponding data, it turns out that they don't have the same size. In general, scRNA assumes that the number of genes/transcripts in the data is the same as in the gene id files. Hence, this is a 1-on-1 matching.