Merge branch 'main' of git.embl.de:grp-arendt/CoFFE into main

README.md

@@ -25,7 +25,11 @@ We used AlphaFold to predict structures for the proteome of _Spongilla lacustris
 - Niko Papadopoulos, Fabian Ruperti, Jacob Musser, and Detlev Arendt wrote the manuscript.
 
 ## License
-MIT, probably?
+GPL 3.0
 
 ## Project status
-ongoing
+Directions pursued currently and in the future:
+- rework code, modularize, apply to additional (non-UniProt) species
+- add sequence profiles comparison (profile-sequence, profile-profile searches)
+- confirm existence of HGT candidates on _Spongilla_ genome, address possible mechanisms
+- apply to proteomics data, other cell types

scripts/align.sh

@@ -11,7 +11,7 @@ module load GCC
 module load bzip2
 module load CUDA
 source ~/.bash_profile
-conda activate /g/arendt/npapadop/repos/condas/maf
+conda activate /g/arendt/npapadop/repos/condas/CoFFE
 
 MMSEQS="/g/arendt/npapadop/repos/MMseqs2/build/bin/mmseqs"
 QUERY="/g/arendt/data/spongilla_singlecell_dataset/spongilla_lacustris_Trinity.fasta.transdecoder_70AA_mediumheader.fasta"
@@ -20,4 +20,4 @@ BASE="/scratch/npapadop/"
 cd "${BASE}"
 colabfold_search "${QUERY}" database/ ./ --mmseqs "${MMSEQS}" --threads 32
 
-module unload
+module unload
\ No newline at end of file

scripts/databases.sh

@@ -69,4 +69,4 @@ if [ ! -f COLABDB_READY ]; then
 fi
 
 module unload aria2
-module unload bzip2
+module unload bzip2
\ No newline at end of file

scripts/predict_structures.sh

@@ -20,7 +20,7 @@ shift
 shift
 unset CONDA_SHLVL
 eval "$(conda shell.bash hook)"
-conda activate /g/arendt/npapadop/repos/condas/milot
+conda activate /g/arendt/npapadop/repos/condas/CoFFE
 export TF_FORCE_UNIFIED_MEMORY=1
 mkdir -p ${OUTPUT}
 colabfold_batch "$INPUT" "$OUTPUT" --stop-at-score 85

scripts/submit_colab.sh

 #!/bin/bash -ex
 
 for i in $(seq 1 32); do
-    sbatch /g/arendt/npapadop/repos/spongfold/milot.sh /scratch/npapadop/msas/batch"$i" /scratch/npapadop/res_batch"$i"
+    sbatch /g/arendt/npapadop/repos/spongfold/scripts/predict_structures.sh /scratch/npapadop/msas/batch"$i" /scratch/npapadop/res_batch"$i"
 done