moca_blue

[2024-03-13]
Welcome the the moca_blue suite!
from Simon M. Zumkeller
RStudio
2022.07.2 Build 576

This is a tool-box for the analyses of DNA motifs
that have been derived from deep-learning model features extraction software 'deepCRE'.
https://github.com/NAMlab/DeepCRE
moca_blue is currently in development.

################## ---------- Workflow example -------- ###################################
#moca_blue has a defined directory structure that needs to be established for proper usage#
mkdir moca_blue
cd mocablue
#Generate folder structure
mkdir 0MOTIFS mo_nom mo_range mo_proj mo_clu ref_seq ..

#Results will be stored in the out directories
mkdir 0MOTIFS mo_nom/out mo_range/out mo_proj/out ..

Assign nomenclature and extract EPMs into jaspar file format

mo_nom Extract motifs from TF-MoDisco hdf5 files, assigns nomenclature and produces weblogos.
Currently, there are three versions of the same script that can be used for the extraction
of a given format of weight matrix.
PFM - positional frequence matrix
PWM - positional weight matrix (best for clustering/comparison)
CWM - contribution weight matrix (best for mapping)

./mo_nom/get_rdf5_cwms_per_pattern_v1.1.R

#Compare EPMs to JASPAR2020 database
./mo_nom/mo_compare_JASPAR2020_v1.0.R

Compare EPMs to themselves, other files in jaspar format or JASPAR2020 database

mo_clu

Analyse and Edit motif-files stored in jaspar-format here. Results should be stored in the "out" directory.
Generates dendrograms/trees based on similarity-matrix for EPMs and Visual.

./mo_clu/mo_cluster_v2.7.R
./mo_clu/mo_compare_JASPAR2020_v1.0.R

#Visualize EPM clustering results
./mo_clu/mo_tree_viz.SZ.v1.0.R

Extract saliency maps and importace scores

./mo_imp/rdf5_get_epm_contrib_scores.v1.1.R
./mo_imp/mo_imp_scores.v1.1.R

#Visualize saliency maps and importace scores (in development)
./mo_imp/mo_imp_depth_v0.7.R

Extract seqlet occurring ranges, positional preferences of EPMs

mo_range ------------------------

Motifs/ EPMs are not distributed at random in a genome.
To optimize the search for motifs/EPMs in a genome or gene-space, these tools
extract the positionally preferred ranges for each motif/EPM in a hdf5 file.

rdf5_get_seql_per_patternV2.R - Extract a list of seqlets and their positions from the hdf5 file
meta_motif_ranges_characteristics_TSS-TTS.1.1.R - Producee a table from the rdf5_get_seql_per_patternV2.R output
that provides the gene-space statistics for each motif/seqlet in reference to transcription start and stop sites (TSS, TTS)

./mo_ran/rdf5_get_seql_per_patternV2.1.R
./mo_ran/epm_occurence_ranges_TSS-TTS.1.6.R

mo_nom/rdf5_get_cwms_per_pattern.v1.0.R
./mo_clu/Mo_cluster_v2.0.R

./mo_range/rdf5_get_seql_patternV2.1.R
./mo_range/meta_motif_ranges_characteristics_TSS-TTS.1.4.R

Map motifs.jaspar to reference genome using BLAMM (https://github.com/biointec/blamm)

#Follow BLAMM installation guide
mv ./blamm_meV1.0.sh ../blamm-master/build/blamm_meV1.0.sh
#Edit sequences.mf file to specify target file for EPM search; e.g. ./ref_seq/file.fas
./blamm-master/build/blamm_meV1.0.sh
cp ./blamm-master/build/outPROJECT ./moca_blue/mo_proj/outPROJECT

Analyze and Visualize the EPM search results

mo_proj -------------------------
After searching for motif matches within a reference sequence, these results can be tested and further characterized using the occ_filter, mo_feat-filter, mo_feature_tester and mo_predictability scripts.
#When the occurence files are to large please use the split_files.sh to split them #########

./mo_proj/occ_filter_v1.1R
./mo_proj/mo_feat-filter.v3.4.R
./mo_proj/mo_feature_tester.v1.0.R
./mo_proj/mo_predictabilityV1.5.R
./mo_proj/mo_check_mapping-performance_V1.7.R
./mo_proj/mo_genotype_variance.v1.4.R

##############################################################################################
ref_seq -------------------------
Store genome data like fasta, gff and many more here for INPUT.
0MODELS (formerly 0MOTIFS)
Store deepCRE output here.
##############################################################################################

---------- Overview --------
recommended directory structure
Please find more detailed descriptions
of the directories and the code in the files itself.

This is a pipeline of consecutive operations that can be and will be availabe here.

INPUT DIRECTORY /ref_seq /0MODELS
- fastas [deepCRE input] - HDF5/H5.file [deepCRE output]
- gffs |___________________
- meta-data | | |
START DIRECTORY | /mo_nom /mo_range /mo_imp
output | - get motif patterns - get motif meta-data - get model meta data
| - motif annotation | |
| - motif modification | |
|||__|
| |
/mo_clu | MAPPING to reference (external)
- analyze motifs | use e.g. "blamm
- compare/cluster | (https://github.com/biointec/blamm)
| cp occurences.txt [results] /mo_proj
||
|
/mo_proj
- filter for meaningful matches
- interpret model predictions
- gene annotation
- module generation
- get bed files for visualisation in JBROWSE

###################################################################################################

mo_old --------------------------- Old and outdated scripts used for the moca_blue suite are stored here.