xinliu005
commented
6 years ago Requirement: 1) Build C. hominis reference genome by using 8 chromosome assembly sequences: LN877947-LN877954 2) Build C. parvum reference genome by using 8 chromosome assembly sequences: CM000429-CM000436 3) a) Update reference-mapping.py to run bwa as the mapping software. The input data shall be either one fastq file or two fastq files, the reference genome fasta file*, and the genome name mapped to. If no reference genome provided, the above genomes 1) or 2) will be used as default. The output file will be SAM file created by bwa.
b) The script need to get a prefix to add to all the file names, lets say if the user wish to have ERR1234 as the prefix all the files that is created by the script shall start by ERR1234_
c) Also use the name of the mapping program in all the file names, so the naming structure for above examples would be:ERR1234bwa*
d) The genome name 'ch' or 'cp' will be added to the prefix, so the example name will be:ERR1234_bwach*
e) The run_type 'single' or 'paired' will be added to the prefix, so the example name will be:ERR1234_bwa_chpaired*
- If the user provide the reference genome by themselves, the following genome index files (.bwt, .pac, .ann, .amb, .sa) should be also provided in the same directory of the genome fasta file. The following command can be used to create the above files: bwa index -a bwtsw .fasta
Question 1: Do we want to map the contigs or reads against the reference genomes? Simone: Yes, I think that this is necessary for a number of things, including identification of core and accessory genes,
Nima: OK great, shall we map the contigs we get from the assembly stage against the reference genome? Or map the raw reads against the reference genome? I guess you want to map reads! Simone: well, we will surely map the reads, but to look for rearrangements, one can use MAUVE, and I think contigs
Question 2: What are the reference genomes? We need ENA accessions. C. parvum:? Study: PRJNA144 ASM16534v1 (IOWA strain) Great this has the full assembly, contigs and chromosomes. http://www.ebi.ac.uk/ena/data/view/PRJNA144 http://www.ebi.ac.uk/ena/data/view/GCA_000165345
This has WGS assembly: http://www.ebi.ac.uk/ena/data/view/GCA_001593465 And two sets of reads: http://www.ebi.ac.uk/ena/data/view/SRR1015747-SRR1015748 If we want to use the reads here we may need to investigate the difference between read sets. We should also check how they have assembled it. Does C. hominis also have 8 chromosome like C. parvum?
Simone: yes, C. hominis has 8 chromosomes, too. I will double check what we should retrieve for hominis.
Question 3: Is there any other parasite reference genomes that we wish to include? Please list them and also provide the accession numbers for them. Simone: No, I think we should include the two above because are the best reconstruction of the genome, and we also have fresh transcriptomics data. Other genomes are only drafted. Nima: Great thanks. Question 4: What are the standard parameters that you wish to extract from this (A part from coverage)? Simone: I think mapping is useful to estimate whether there is a bias in the sequencing experiments, i.e., if there are regions that are under-represented. Mapping data can also be used to visualize rearrangements. At least in the commercial software we have(CLC Genomics Workbench), mapping is used to call SNPs, because if one has all the sequences from one isolate mapped to a reference genome, then it is possible to exclude SNPs by selecting initial options (e.g., a minimum and/or maximum coverage,quality of the nucleotide and surrounding nucleotides, accounting for bias in the ratio forward/reverse reads, ecc)
Nima: Yes that’s correct, it is needed for SNPs analysis.