shamir-lab / adaorder

AdaOrder

This implementation of AdaOrder corresponds to the following paper: .

To cite this paper use the following form:

In orders_minimizer_7.zip are pre computed orders by AdaOrder, on 4 datasets used in the paper with k=28,55.

Usage

First, compile the source code using your compiler to "AdaOrder.jar".

AdaOrder generates 2 files in the working directory: freq.txt, ranks.txt. ranks.txt is the file with the generated minimizer order [Algorithm 1 in our paper]. freq.txt is the file with the estimated sizes if minimizers bins [Algorithm 2 in our paper]. In both files: line i corresponds to the minimizer with value i using Natural, and its value is its (rank \ bin size) among all m-mers.

    java -classpath "AdaOrder.jar" dumbo.OrderingOptimizer  -in <input-file>  <parameters>;      

AdaOrder can be controlled by several command line options and flags.

Option	Description	Default
‑k <int>	Set the length of k-mers.	60
‑m <int>	Set the length m of minimizers.	7
‑R <int>	Set the number of AdaOrder runs.	1000
‑N <int>	Set the number of samples per round of AdaOrder.	100000
‑p <float>	Set the penalty factor of AdaOrder.	0.01
‑in <path>	Dataset to generate AdaOrder for.

DGerbil

DGerbil is a modification of Gerbil, a memory efficient k-mer counter, that uses AdaOrder instaed of signature, for improved memory usage. Its source code is found at https://github.com/Shamir-Lab/DGerbil.