Mar 9, 2020

Written by Joshua Dempster

Introducing the CDS Blog

Welcome

Jul 25, 2023

Written by Joshua Dempster

Evaluating gene relatedness in CRISPR

One use of large CRISPR datasets is to find functional relationships using “guilt-by-association”. However, measuring how well a dataset does this turns out to be surprisingly tricky.

May 1, 2021

Written by Joshua Dempster

Understanding the differences between CERES and Chronos DepMap data

A deeper exploration of how and why gene effect scores from CERES and Chronos might differ.

Aug 26, 2020

Written by William Colgan

Working with MIX-Seq data

We recently published a new method called MIX-Seq. This post will get you started analyzing MIX-Seq data using R and Seurat

Jul 2, 2020

Written by Mariya Kazachkova

The MON artifact

Identification and consequences of an artifact in Project Achilles

Jun 26, 2020

Written by Joshua Dempster

Assessing Confidence in Achilles Gene Profiles

A long post which contains both an explanation of how we calculate gene confidence scores together with the code we use to do so.

Apr 23, 2020

Written by Philip Montgomery

Speeding up calculation of Pearson correlation for matrices with missing data

It's easy enough to calculate pearson correlation efficiently using numpy, but what if you have missing values in your data?

Apr 7, 2020

Written by Joshua Dempster

Library permutation leads to overly optimistic p-values in CRISPR screens

Permuting reagent labels is a common approach to generate null distributions in CRISPR screens. This post discusses the problems with this method and suggests an alternative.

Mar 9, 2020

Written by Mariya Kazachkova

Genes with no unique* sgRNAs are being dropped

Problem with and solution for multi-gene targeting sgRNAs

Mar 9, 2020

Written by Joshua Dempster

When not to use Spearman correlations

Spearman correlations are frequently used in computational biology, but for a common class of biological relationships Pearson is a better choice.