Song Lab @ Cal

tape

Tasks Assessing Protein Embeddings (TAPE), a set of five biologically relevant semi-supervised learning tasks spread across different domains of protein biology.

gpn

Genomic Pre-trained Network

tape-neurips2019

Tasks Assessing Protein Embeddings (TAPE), a set of five biologically relevant semi-supervised learning tasks spread across different domains of protein biology. (DEPRECATED)

factored-attention

This repository contains code for reproducing results in our paper Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention

mogwai

rna-sieve

A library for the deconvolution of bulk cell samples using single-cell RNA expression data.

CPT

Cross-protein transfer learning for variant effect prediction

scquint

swga2

CherryML

Scalable Maximum Likelihood Estimation of Phylogenetic Models

scquint-analysis

contact-geometry

MultiCluster

Software for three-way clustering of multi-tissue multi-individual gene expression data using semi-nonnegative tensor decomposition

MOCHIS

One- and two-sample tests against general families of alternatives

flinty

A Simple and Flexible Test of Sample Exchangeability with Applications to Statistical Genomics

HiDENSEC

Code for Tracing cancer evolution and heterogeneity using Hi-C

EGGTART

Extensive GUI gives TASEP realization in real-time.

slc22a5

Variant effect prediction for SLC22A5 transporter gene using Potts models

l-TASEP

SGDP_IGHV_TRBV

Clustered and filtered reads after mapping for the SImons Diversity Genome Project.

epigenome_editing_2023

Code for training and analyzing models from the epigenome editing paper Batra et al.

fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

forward-equivalent-trees

Efficiently simulating multi-type birth-death processes via forward-equivalent parameter mapping

slip

SLIP is a sandbox environment for engineering protein sequences with synthetic fitness functions.

songlab-cal.github.io

StableFM

Data and scripts for fine-mapping stability project

StratPGS

Performance Inflation and Perturbation Sensitivity of PGSs