Draft Paper of "Purifying Selection Maintains Conserved Noncoding Elements in Drosophila (Chapter #3)"

[MManee]

TODO

• [x] convert chapter 3 into a paper manuscript
• [x] make a new directory and creating a new issue in my bergmanlab github repo
• [x] make a new .tex document and pasting in the tex from chapter 3
• [x] update paths to figures/files, etc.
• [x] perform a code review and replication run before submission

[MManee]

TODO

• [x] front page - article title, authors and their affiliation(s)
• [x] abstract and keywords
• [x] introduction
• [x] materials and methods
• [x] results
• [x] discussion
• [x] minor sections - author Contributions, acknowledgment and etc

[MManee]

• in latex file -> discussion section -> paragraph 2

  regions which are known to target promoter regions \citep{spradling_drosophila_2011}. However, large

• the reference \citep{spradling_drosophila_2011} produces this error Misplaced alignment tab character &.

[MManee]

• the code does not generate dm3_exons.txt file, it should be provided with the code files
• the code should generate high_recomb_coords.bed file (recommended)
  • so far, the code does not generate it

[MManee]

• I updated perl script (_get_observed_simulated_tes.pl_), I am testing the script now
  • almost every dataset can be downloaded automatically excepting three datasets:
  • dm3 exons dataset, coordinates of high recombination dataset and _ngs_temapper TE strain counts dataset
  • I need to look for direct links to download them to be included in perl script
• for TEMP TE detection method, I used the following steps to generate the TE dataset:
  • TEMP set was downloaded from: https://zlab.umassmed.edu/TEMP/TEMP_resources/DGRP_53lines_TE_polymorphisms.tar.gz
  • predictions with 1p1 status were extracted for each strain. 1p1 status means that the detected insertion is supported by at least one supporting read at each end of the TE (Total TEs: 18441)
  • predictions for different strains within TE families were then merged to create a non-redundant set of insertion sites using BEDTools merge with -o and -c options bedtools merge -i temp_te_sorted.bed -c 1 -o count > temp_te_merged.bed (Total TEs: 13593)
    • the -o and -c options were used together to report the count of intervals that merged in each resulting interval
  • only 2L, 2R, 3L, 3R, 4 and X chromosomes were included (Final Total TEs: 12500)
• the next step is to update R script (te_pipeline.r) to merge plots of _ngs_temapper and TEMP results

[MManee]

I originally downloaded the coding-exon only data file (dm3 exons) I am using now from the UCSC table browser

[MManee]

TODO

• [x] dig out the citation for where I can get the coordinates of high recombination regions
• [x] make perl script creates file of high recombination region coordinates
• [x] automatically download dm3 exon data from here: http://hgdownload.soe.ucsc.edu/goldenPath/dm3/database/flyBaseGene.txt.gz. But I need to parse the exons from the exon start and stop fields, see documentation here: http://hgdownload.soe.ucsc.edu/goldenPath/dm3/database/flyBaseGene.sql

[MManee]

• I parsed exons start and stop fields from flyBaseGene file:
  • I loaded the exon tracks into genome browser to check them: View Exons on Genome Browser

[MManee]

• The coordinates I am using to define high recombination regions are described in Bergman et al 2006 which was on release 4, lifted over to release 5
• it looks like there are 2 relevant papers by Cridland et al (that don't do anything I do :) which use a similar set of coordinates to restrict their analysis:
  • https://academic.oup.com/mbe/article/30/10/2311/1246156
  • http://www.genetics.org/content/199/1/85.long

    chrX:300000–20800000
    chr2L:200000–20100000
    chr3L:100000–21900000
    chr2R:2300000–21000000
    chr3R:600000–27800000

• here my current coordinates for comparison:

  chrX    1   20429372
  chr2L   1   19673576
  chr3L   1   20604281
  chr2R   3068177 21146708
  chr3R   2811816 27905053

• The Cridland coordinates are trim a little from the telomeric parts of the chromsomes where reference TE density is high, but also extend further into the centromeric parts of the chromosomes. It would be worth seeing if I get the same overall results using the Cridland coordinates, since it would allow me to cite their work, more directly compare with their results and possibly increase our sample size of TE insertions.
• I have to run with the Cridland coordinates and let Casey see the results

[MManee]

TODO

• [x] run the pipeline with the Cridland coordinates and post the results
• [x] the sum of y-axis for the allele frequency plots should not go above 1, and change the color of the different categories

[MManee]

• The paper was accepted and published on 29 May 2018: Go to the paper