Preprocessing single-cell BCR sequencing and transcriptomic data
This blog post covers the basic commands, data cleaning and analysis angles for scBCR datasets and transcriptomic studies.
PhD in Bioinformatics, Scientist in Antibody Development
This blog post covers the basic commands, data cleaning and analysis angles for scBCR datasets and transcriptomic studies.
Over the break, I was looking at competitive coding and had a proper look at different types of algorithms. I was presented with a task to find the shortest path to traverse a graph – a classical algorithmic problem that is surprisingly common in both Computational Chemistry and Biology. In this blog post, I am attempting to explain to my future self how to do these searches using efficient algorithms.
I was testing out all the possible ways to improve the performance of hashing and matching for the Ligity paper. Here is one of the cases where I benchmarked a bloom filter:
N=5
num_jobs="\j"
for F in ./*.txt
do
if [ -f ${F%.txt}.csv ]; then
continue
fi
while (( ${num_jobs@P} >= N )); do
wait -n
done
echo ${F%.pip}.ht
../main.o -i ${F} -o ${F%.txt}.csv &
done Take home: use numpy instead of pandas if you have numerous calculations.
Source: https://engineering.upside.com/a-beginners-guide-to-optimizing-pandas-code-for-speed-c09ef2c6a4d6