- Python 66.5%
- R 22.9%
- Perl 6.3%
- Shell 4.3%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| ancient_dna_cpp_tools@63ec7a40d3 | ||
| scripts_for_SediQuest | ||
| .gitmodules | ||
| check_mapped_ref.py | ||
| config_example.yaml | ||
| coverage_plot.png | ||
| Lib.S.9756_kraken_fam_spe_byburden.pdf | ||
| Lib.S.9756_kraken_order_byburden.pdf | ||
| LICENSE.txt | ||
| README.md | ||
| SediQuest.smk | ||
| SediQuest_v2.png | ||
This pipeline is a friendly-user implementation of the Vernot et al. (2021) workflow. It aims to extract human reads from capture data to perform further analysis. The set of sites used to capture the DNA is called a probeset here.
The aim of this pipeline is to filter out mammalian contamination. To achieve this, a Kraken step is performed at stages X and X of the pipeline, and three sensitivity filtering parameters are available:
- Low sensitivity:
score_b: "ALL"
score_n: "ALL"
filter: "LOW"
- Middle sensitivity
score_b: [x]
score_n: "ALL"
filter: "LOW"
x being a number you will chose according to the coverage plot (see the output section)
- High sensitivity
score_b: [x]
score_n: "ALL"
filter: "HIGH"
This filtering is a combined filter of mammalian diversity score and kraken to save as much data as you can in a case of a highly faunal contaminated sample
Requirements
- samtools 1.18
- bedtools v2.25.0
- Python 3.6.15
Required Files
config.yaml– Configuration file with parameters, see config_example.yaml to have more information.
Please be aware the pipeline by default should first be run with this parameters:
score_b: "ALL"
score_n: "ALL"
filter: "LOW"
probeset.csv– Table with information about the nuclear probes each sample was captured with. This table should include:- A modified reference genome where you replaced the target sites by a "third" base (see Vernot et al. 2021).
- An optional BED file with mammalian diversity scores for filtering (useful if you set score_b to something else then ALL)
- A BED file with target regions
| probeset | path_to_ref | path_to_bed | path_to_control |
|---|---|---|---|
| 1240k | whole_genome_modified_1240k.fa | mammalian_diversity_score_1240k.bed | 1240k.bed |
samples.csv– Table with sample information. Must include at least:- LibraryID
- Probeset
| probeset_to_lib | indexlibid |
|---|---|
| 1240k | Lib.1608 |
Time to play with your data
Before – Verify Mapping
To ensure your genome is mapped to the modified reference genome corresponding run:
python check_reference_mapping.py
Run the pipeline
snakemake run_pipeline --snakefile SediQuest.smk --configfile config_example.yaml --cores 25
After - Check the output
Plots and Tables are available for you to look at, here is a bried description of them:
- Summary table
| IndexLibID | N_score | MD_score | probeset | split | mapped | unique | target | kraken_target | deam | kraken_deam | 5CT | 3CT | 5CT_95CI | 3CT_95CI | cond5CT | cond3CT | cond5CT_95CI | cond3CT_95CI | average_dup | SNPs_count_target | SNPs_count_deam | Contamination | Contamination_err_estimate |
|---|
- Overall coverage plot
This plot is an example, generated from a deer bone and a human bone for you to investigate if your sediment sample look like it has faunal contamination in it or not.
- Kraken coverage plot
This plot is an example from a sediment sample

