This is package is the library design tool for the Yesselman Lab (https://yesselmanlab.com/). This tool takes a set of RNA sequences and finalizes them so they can be ordered from IDT, agilent or Twist or others. There are options to add unique barcodes to increase the diversity of libraries. There are also several checks to stop ordering libraries that will not work.
All libraries are assumed to be RNA and a T7 promoter (TTCTAATACGACTCACTA) will be added to the 5' end of each sequence. So that the library will be able to be transcribed into RNA
It is highly recommended to install with requirements in a conda environment
conda create --name py3 python=3.8
conda activate py3 This package is not meant to be on pypi atm so the option is local install
git clone https://github.com/jyesselm/rna_lib_design.git
cd rna_lib_design
pip install .This will create a executable in your path called rld
the rld command line tool has 5 sub commands. list barcode barcode2 add-common and edit-distance
rld --help
Usage: rld [OPTIONS] COMMAND [ARGS]...
This is a simple tool to generate barcodes for a library of RNAs to be ordered
as an DNA oligo library.
Options:
--help Show this message and exit.
Commands:
add-common add common p5/p3 sequences
barcode adds a single barcode
barcode2 adds two barcodes
edit-distance compute edit distance of library
list lists resources availablelist displays the available barcodes and common sequences that can be used in the other sub commands
edit-distance will compute the edit distance of the library
The other sub commands are different methods of finalizing RNA libraries so they can be ordered as DNA oligo pools.
All libraries should be supplied as csvs. Lets start with a simple example such as
$ cat test/resources/libs/simple.csv
sequence
GGGGGGAAAACCCCCC
CCAAAACCCCUUUUGGAll parameters to be used by each sub command is are stored in parameter files. You can see all of these files in the rna_lib_design/resources/presets directory. The parameters are stored in a yaml file. These parameters are validated by the jsonschema package. The schema for the parameters is stored in rna_lib_design/resources/schemas/.
Here is a breakdown of the parameters.
The build string defines what segments of RNA are we going to add to each sequence. For example.
build_str: P5-P5EXT-SOI-P3EXT-P3Each build string MUST contain SOI or sequence of interest. All other are optional but if they appear in the build string they must also be included in segments which we will discuss in a second. The position of each segment in the build string is important. The order of the segments will be the order they are added to the sequence of interest. So for this build string the segment of P5EXT will be added immediately to the 5' end of SOI and P3EXT will be added immediately to the 3' end of SOI. P5 will be added to the 5' end after P5EXT and etc.
Each named segment in the build_str must be defined in another section called segments here each name must have its own parameters defined. There are many possibilites. using the name option defines a set sequence that will be added to each sequence of interest in the library. You can see the corresponding values foreahc name in the rna_lib_design/resources/named_eqs directory.
segments:
P5:
name: "uucg_p5_rev_primer"
P3:
name: "rt_tail"
P3EXT:
sequence: "AC"
structure: ".."
P5EXT:
sequence: ""
structure: ""We will use this file as an example for the rest of the documentation.
The add-common sub command will add 5' and 3' sequences to each sequence in a csv file.
rld add-common --help
Usage: rld add-common [OPTIONS] CSV
add common p5/p3 sequences
Main options:
These are the main options for generating a library
-t, --btype TEXT what type of barcode to use see full list in
resources/presets
--param-file PATH supply a new param file to override specific
present or to manuallydetermine each option
-o, --output TEXT the path to save results to
-p, --num-processes INTEGER number of processes to run simultaneously
--debug turn on debug logging for the application
--skip-edit-dist skip the edit distance calculation
--trim-p5 INTEGER trim sequence at 5' end by this length
--trim-p3 INTEGER trim sequence at 3' end by this length
Other options:
--help Show this message and exit.Lets add the standard p5/p3 sequences to our simple library. If no additional options are supplied the standard preset will be used. The standard preset is defined in rna_lib_design/resources/presets/add_common_standard.yml and can be overridden with the --param-file option.
All the parameters are displayed in the log output. Every single one can be changed. The parameters are stored in a yaml file and can be edited by hand. The parameters are also stored in the results directory for future reference.
$ rld add-common test/resources/libs/simple.csv
INFO RLD.CLI Using csv: test/resources/libs/simple.csv
INFO RLD.CLI Using output dir: results
INFO RLD.CLI Copying test/resources/libs/simple.csv to results/input.csv
INFO RLD.CLI csv has 2 sequences
INFO RLD.CLI Writing parameters to results/params.yml
INFO RLD.CLI No preset or param file supplied, using standard preset
INFO RLD.CLI Using parameters:
{
"build_str": "P5-P5EXT-SOI-P3EXT-P3",
"segments": {
"P5": {
"name": "uucg_p5_rev_primer"
},
"P3": {
"name": "rt_tail"
},
"P3EXT": {
"sequence": "AC",
"structure": ".."
},
"P5EXT": {
"sequence": "",
"structure": ""
}
},
"design_opts": {
"increase_ens_defect": 2.0,
"max_ens_defect": 5.0,
"max_attempts": 10,
"max_solutions": 10,
"score_method": "increase",
"allowed_ss_mismatch": 2,
"allowed_ss_mismatch_barcodes": 2
}
}
INFO RLD.DESIGN starting design
INFO RLD.DESIGN no 'name' column was in dataframe - adding one
INFO RLD.DESIGN running on single core
INFO RLD.DESIGN no 'structure' column folding it now
INFO RLD.SSET P5 is using a named sequence/structure: uucg_p5_rev_primer
INFO RLD.SSET P5 -> SequenceStructure(sequence='GGAACAGCACUUCGGUGCAAA', structure='......((((....))))...')
INFO RLD.SSET P3 is using a named sequence/structure: rt_tail
INFO RLD.SSET P3 -> SequenceStructure(sequence='AAAGAAACAACAACAACAAC', structure='....................')
INFO RLD.CLI no sequences discarded
INFO RLD.DESIGN results/results-all.csv contains all information generated from run
INFO RLD.DESIGN results/results-rna.csv contains only information related to the RNA sequence
INFO RLD.DESIGN p5 seq -> SequenceInfo(name='uucg_p5_rev_primer', sequence='GGAACAGCACUUCGGUGCAAA', code='P0058')
INFO RLD.CLI the edit distance of lib is: 12.0
get docker working for automated testing allow vienna rna to be installed from source so it works on all operating systems new types of barcodes? triple barcoding? double barcodes with single strands?