User Manual for SweepFinder2
Contents
1. black dots in Figure 1A below SweepFinder2 lr 100 BGS noSweep SF 65 invarl Neutral _Background sfs invarl BGS noSweep Rec map 65 invarl Out txt Here is an example of using SweepFinder2 to identify selective sweeps under simulations with only background selection but while using input recombination and B value maps The output of this analysis will produce output used for the red dots in Figure 1A below SweepFinder2 lrb 100 BGS noSweep SF 65 invarl Neutral Background sfs invarl BGS noSweep Rec_map 65 invarl BGS noSweep Bval_map 65 invarl 250 250 2000 Out txt Here is an example of using SweepFinder2 to identify selective sweeps under simulations with both background and positive selection but while using an input recombination map but not an input B value map The output of this analysis will produce output used for the black dots in Figure 1B below SweepFinder2 lr 100 BGS Sweep SF 84 invarl Neutral Background sfs invarl 8 BGS Sweep Rec_map 84 invarl Out txt Here is an example of using SweepFinder2 to identify selective sweeps under simulations with both background and positive selection but while using input recombination and B value maps The output of this analysis will produce output used for the red dots in Figure 1B below SweepFinder2 lrb 100 BGS Sweep SF 84 invarl Neutral Background sfs invarl BGS Sweep Rec map 84 invarl BGS Sweep Bval map 84 invarl 250 250 20002. contained in FreqFile k k 1 2 22 4 1 Compute empirical frequency spectrum To compute the empirical frequency spectrum use the f option identical to the command from the original version of SweepFinder The command is SweepFinder2 f CombinedFreqFile SpectFile where CombinedFreqFile is an allele frequency input file combined across all chromosomes in the analysis to get a genome wide estimate and Spect File is the name of a file where the results will be printed 5 Scanning for selective sweeps 5 1 Scan for selective sweeps To perform a scan for selective sweeps with the original method of Nielsen et al 2005 use the s option The command to perform this scan is SweepFinder2 s G FreqFile OutFile where G is a user defined number of grid points G test sites are equally spaced across the genomic region spanned by the positions in FreqFile to compute the test statistic FreqFile is the allele frequency input file and Out File is the name of a file where the results will be printed Here FreqF ile would be for a specific chromosome or region of the genome rather than combined across all chromosomes Sometimes it is more convenient to set the spacing between grid points rather than the number of grid points The user may specify the approximate desired spacing between test sites using the sg option The command to perform this scan is SweepFinder2 sg g FreqFile OutFile where g is a user defined s
3. 8 1 463000 0 94 0 The first line of the example input file displays the header which must be identical to this example The next four rows display allele frequency data for four positions on a chromosome positions 460000 460010 460210 and 463000 Each row indicates the number of derived alleles observed and the total number of alleles observed at the given chromosomal position At position 460000 9 derived alleles were observed out of 100 total alleles 50 diploid individuals leading to an observed polymorphism in the sample At position 460010 100 derived alleles were observed out of 100 total alleles 50 diploid individuals leading to an observed substitution in the sample At position 460210 30 alleles of one type were observed out of 78 total alleles 39 diploid individuals leading to another observed polymorphism in the sample Note that at this row we set folded to 1 because we were not sure whether the allele was derived or ancestral At position 463000 no derived alleles were observed out of 94 total alleles 47 diploid individuals leading to an observed monomorphic site that is identical to the outgroup in the sample If the true sample size was 50 diploid individuals then positions 460210 and 463000 would be genomic positions with missing data in the sample 3 2 Recombination file The recombination rate input file is tab delimited and contains a header Each row represents the recombination rate in centiMorgans cM between th
4. Out txt A120 Without B value map i y 80 With B value map i 5 e a O 40 ia c eso gee TE Oe PEONIO IEE IEE TEE ETETEA l l T ooto 900000o0000o E B120 x Without B value map Loooto AS i g 80 With B value map i O 4 O 40 i i B Feoqneece corretoceese i 0 Le0000000 000t0000000p 0000000000000 l l TTT TOTTALY eo ee 0 10 20 30 40 50 60 70 80 90 100 Position kb Figure 1 Results from applying SweepFinder2 to data simulated with background selection A B Composite likelihood ratio test statistics as a function of position along a sequence without A and with B a fixed selective sweep in the center of the sequence The gray region represents a reduction in recombination rate by two orders of magnitude Including the B value map decreases false inferences of positive selection A yet still can identify positively selected alleles in regions with background selection B 7 References CD Huber M DeGiorgio Hellmann R Nielsen 2015 Detecting recent selective sweeps while controlling for mutation rate and background selection Mol Ecol doi 10 111 mec 13351 M DeGiorgio CD Huber MJ Hubisz Hellmann R Nielsen SweepFinder2 increased robustness and flexibility Submitted R Nielsen S Williamson Y Kim MJ Hubisz AG Clark C Bustamante 2005 Genomic scans for selective sweeps using SNP data Genome Res 15 156 1575
5. The positions in the user defined grid file should be spanned by the range of positions in the allele frequency input file Only those positions in the user defined grid file will have a selective sweep computed That is providing a user defined grid file overrides the uniform grid option that is default to SweepFinder2 An example user defined grid input file 460000 463000 The first four rows indicate that a test for selective sweeps will be computed at positions 460000 460010 460210 and 463000 4 Helper file useful for genome wide scans of selective sweeps In many circumstances it is desirable to calculate the frequency spectrum across the whole genome and to use this frequency spectrum as the allele frequency distribution under the null hypothesis of neutrality It is required that the user first combine their allele frequency files into a single allele frequency file with exactly the same format as the example in section 3 1 and we will refer to this file as CombinedFreqFile The reason to create this combined allele frequency file is to generate genome wide estimates of the empirical frequency spectrum As an example suppose we have data from each of the 22 human autosomes and each chromosome k has its own allele frequency file called FreqFile k k 1 2 22 The CombinedFreqFile would have all of the data contained in FreqFile k k 1 2 22 in one file There would be a line in CombinedFreqFile for each of the data lines
6. User Manual for SweepFinder2 Michael DeGiorgio Christian D Huber Melissa J Hubisz Ines Hellmann Rasmus Nielsen October 6 2015 1 Introduction SweepFinder2 is a program to perform genome wide scans of selective sweeps using the spatial distribution of allele frequencies and selective constraint in the genome The current version of the software implements the test for sweeps in the context of background selection of Huber et a 2015 and DeGiorgio et al submitted If you identify any bugs or issues with the software then please contact Michael DeGiorgio at mxd60 psu edu to report the issue If you use this software then please cite it as M DeGiorgio CD Huber MJ Hubisz Hellmann R Nielsen SweepFinder2 increased robustness and flexibility Submitted 2 Installation SweepFinder2 should run on any UNIX system On the command line enter gt tar xzvf SF2 tar gz gt cd SF2 gt make The first command decompresses the tar gz file the second command switches to the decompressed directory and the make command compiles SweepFinder2 The executable will be located in the SF2 decompressed directory 3 Input file format Note If you have used a Windows operating system to generate the following input files then you will likely need to run the UNIX command dos2unix on each of the files before using them as input in SweepFinder2 3 1 Allele frequency file The allele frequency input file is tab delimited and contain
7. e Every position and only those positions in the allele frequency input file should be included in the respective B value input file There should be a separate B value file for each chromosome when performing a scan for selective sweeps At each row the first column is the position on the chromosome and the second column is the B value An example B value input file matching the above example allele frequency input file position bvalue 460000 0 2 460010 0 5 460210 0 95 463000 1 0 The first line of the example input file displays the header which must be identical to this example for every B value input file The next four rows display B values for four positions on a chromosome positions 460000 460010 460210 3 and 463000 At position 460000 the B value is 0 2 indicating a strong reduction in diversity is expected due to background selection The B value at position 460010 is 0 5 indicating that background selection reduces diversity by 50 at that site The B value at position 460210 is 0 95 indicating that background selection had minimal impact on diversity at this site Finally the B value at position 463000 is 1 0 indicating that background selection has no impact on diversity at this site 3 4 User defined grid file The user defined grid input file has a simple format with a single position on each row there is no header Each position will specify a genomic location at which the test statistic will be computed
8. e position in the genome and the previous position in the file and the rows are ordered by increasing position along the chromosome Every position and only those positions in the allele frequency input file should be included in the respective recombination input file There should be a separate recombination rate file for each chromosome when performing a scan for selective sweeps At each row the first column is the position on the chromosome and the second column is the recombination rate in cM For the first position in the input file the recombination rate is 0 An example recombination input file matching the above example allele frequency input file position rate 460000 0 0 460010 0 0001 460210 0 002 463000 0 0279 The first line of the example input file displays the header which must be identical to this example for every recombination input file The next four rows display recombination rates in cM for four positions on a chromosome positions 460000 460010 460210 and 463000 At position 460000 the rate is 0 because it is the first position in the file The rate between positions 460000 and 460010 is 0 0001 cM between positions 460010 and 460210 is 0 002 cM and between positions 460210 and 463000 is 0 0279 cM 3 3 B value file The B value input file is tab delimited and contains a header Each row represents the B value at the position in the genome and the rows are ordered by increasing position along the chromosom
9. ical spectrum recombination map and B value map To perform a scan for selective sweeps with a pre computed empirical frequency spectrum and a recombination map use the 1rb option The command to perform this scan is SweepFinder2 lrb G FreqFile SpectFile RecFile BValFile N1 N2 T OutFile where G is a user defined number of grid points G test sites are equally spaced across the genomic region spanned by the positions in FreqFile to compute the test statistic FreqFile is the allele frequency input file RecFile is the respective recombination rate file Bval File is the respective B value file N1 is current ingroup effective population size N2 is ancestral effective population size T is the divergence time in generations between the ingroup and the outgroup SpectFile is an input file containing the empirical derived allele frequency spectrum calculated using the f option in section 4 1 and OutFile is the name of a file where the results will be printed Here FreqFile RecFile BValFile would be for a specific chromosome or region of the genome rather than combined across all chromosomes Sometimes it is more convenient to set the spacing between grid points rather than the number of grid points The user may specify the approximate desired spacing between test sites using the 1 rbg option The command to perform this scan is SweepFinder2 lrbg g FreqFile SpectFile RecFile BValFile N1 N2 T OutFile where g is a
10. ile would be for a specific chromosome or region of the genome rather than combined across all chromosomes Sometimes it is more convenient to set the spacing between grid points rather than the number of grid points The user may specify the approximate desired spacing between test sites using the 1 rg option The command to perform this scan is SweepFinder2 lrg g FreqFile SpectFile RecFile OutFile where g is a user defined space between grid points For example if the user desired a test site approximately every one kilobase then g 1000 representing 1000 nucleotides Further it can often be useful to use a custom grid of test sites rather than a uniform grid The user may specify this custom grid using the 1 ru option The command to perform this scan is SweepFinder2 lru GridFile FreqFile SpectFile RecFile OutFile where GridFile is a user defined grid input file defined in section 3 4 5 4 Scan for selective sweeps with pre computed empirical spectrum and B value map To perform a scan for selective sweeps with a pre computed empirical frequency spectrum and a recombination map use the 1b option The command to perform this scan is SweepFinder2 1b G FreqFile SpectFile BValFile N1 N2 T OutFile where G is a user defined number of grid points G test sites are equally spaced across the genomic region spanned by the positions in FreqFile to compute the test statistic FreqFile is the allele frequency inpu
11. ired spacing between test sites using the 1g option The command to perform this scan is SweepFinder2 lg g FreqFile SpectFile OutFile where g is a user defined space between grid points For example if the user desired a test site approximately every one kilobase then g 1000 representing 1000 nucleotides Further it can often be useful to use a custom grid of test sites rather than a uniform grid The user may specify this custom grid using the 1u option The command to perform this scan is SweepFinder2 lu GridFile FreqFile SpectFile OutFile where GridFile is a user defined grid input file defined in section 3 4 5 5 3 Scan for selective sweeps with pre computed empirical spectrum and recombination map To perform a scan for selective sweeps with a pre computed empirical frequency spectrum and a recombination map use the 1r option The command to perform this scan is SweepFinder2 lr G FreqFile SpectFile RecFile OutFile where G is a user defined number of grid points G test sites are equally spaced across the genomic region spanned by the positions in FreqFile to compute the test statistic FreqFile is the allele frequency input file RecFile is the respective recombination rate file SpectFile is an input file containing the empirical derived allele frequency spectrum calculated using the f option in section 4 1 and Out File is the name of a file where the results will be printed Here FreqFile and RecF
12. ncies for counts x 0 1 n BGS noSweep Rec map 65 invar0 Recombination map for counts x 1 2 n 1 BGS noSweep Rec map 65 invarl Recombination map for counts x 1 2 n BGS noSweep Rec map 65 invar2 Recombination map for counts x 0 1 n BGS noSweep Bval_map 65 invar0 B value map for counts x 1 2 n 1 BGS noSweep Bval_ map 65 invarl B value map for counts x 1 2 n BGS noSweep Bval_ map 65 invar2 B value map for counts x 0 1 n List of files from simulations with both background and positive selection BGS Sweep SF 84 invar0 Allele frequencies for counts x 1 2 n 1 BGS Sweep SF 84 invarl Allele frequencies for counts x 1 2 n BGS Sweep SF 84 invar2 Allele frequencies for counts x 0 1 n BGS Sweep Rec map 84 invar0 Recombination map for counts x 1 2 n 1 BGS Sweep Rec map 84 invarl Recombination map for counts x 1 2 n BGS Sweep Rec map 84 invar2 Recombination map for counts x 0 1 n BGS Sweep Bval map 84 invar0 B value map for counts x 1 2 n 1 BGS Sweep Bval_ map 84 invarl B value map for counts x 1 2 n BGS Sweep Bval_ map 84 invar2 B value map for counts x 0 1 n Here is an example of using SweepFinder2 to identify selective sweeps under simulations with only background selection but while using an input recombination map but not an input B value map The output of this analysis will produce output used for the
13. pace between grid points For example if the user desired a test site approximately every one kilobase then g 1000 representing 1000 nucleotides Further it can often be useful to use a custom grid of test sites rather than a uniform grid The user may specify this custom grid using the su option The command to perform this scan is SweepFinder2 su GridFile FreqFile OutFile where GridFile is a user defined grid input file defined in section 3 4 5 2 Scan for selective sweeps with pre computed empirical spectrum To perform a scan for selective sweeps with the original method of Nielsen et a 2005 and a pre computed empirical frequency spectrum use the 1 option The command to perform this scan is SweepFinder2 1 G FregqFile SpectFile OutFile where G is a user defined number of grid points G test sites are equally spaced across the genomic region spanned by the positions in FreqF ile to compute the test statistic FreqFile is the allele frequency input file SoectFile is an input file containing the empirical derived allele frequency spectrum calculated using the f option in section 4 1 and OutFile is the name of a file where the results will be printed Here FreqFile would be for a specific chromosome or region of the genome rather than combined across all chromosomes Sometimes it is more convenient to set the spacing between grid points rather than the number of grid points The user may specify the approximate des
14. s a header Each row represents the allele frequency for a position in the genome and the rows are ordered by increasing position along the chromosome There should be a separate allele frequency file for each chromosome when performing a scan for selective sweeps At each row the first column is the position on the chromosome the second column is the allele count x the third column is the sample size n and the fourth column is an indicator as to whether the site has been polarized i e whether it is known that the allele is derived or ancestral If the site is polarized then the entry in the folded column should be O and the entry in the second column should be the derived allele count x If the site is not polarized then the entry in the folded column should be 1 The allele count can take on values x 0 1 n If x 0 and the site is polarized then the site is monomorphic and the allele is identical to the outgroup used to polarize the site If x n and the site is polarized then the site is a substitution monomorphic and different from the outgroup used to polarize the site Otherwise the site is a polymorphism Based on the study of Huber et al 2015 we do not recommend using sites with x 0 when the site is polarized That is we recommend only using polymorphisms and substitutions to scan for sweeps An example input file is say for chromosome 6 position x n folded 460000 9 100 0 460010 100 100 O 460210 30 7
15. t file Bval File is the respective B value file N1 is current ingroup effective population size N2 is ancestral effective population size T is the divergence time in generations between the ingroup and the outgroup Spect File is an input file containing the empirical derived allele frequency spectrum calculated using the option in section 4 1 and Out File is the name of a file where the results will be printed Here FreqFile and BValFile would be for a specific chromosome or region of the genome rather than combined across all chromosomes Sometimes it is more convenient to set the spacing between grid points rather than the number of grid points The user may specify the approximate desired spacing between test sites using the 1bg option The command to perform this scan is SweepFinder2 lbg g FreqFile SpectFile BValFile N1 N2 T OutFile where g is a user defined space between grid points For example if the user desired a test site approximately every one kilobase then g 1000 representing 1000 nucleotides Further it can often be useful to use a custom grid of test sites rather than a uniform grid The user may specify this custom grid using the 1bu option The command to perform this scan is SweepFinder2 lbu GridFile FregFile SpectFile BValFile N1 N2 T OutFile where GridFile is a user defined grid input file defined in section 3 4 5 5 Scan for selective sweeps with pre computed empir
16. user defined space between grid points For example if the user desired a test site approximately every one kilobase then g 1000 representing 1000 nucleotides Further it can often be useful to use a custom grid of test sites rather than a uniform grid The user may specify this custom grid using the 1 rbu option The command to perform this scan is SweepFinder2 lrbu GridFile FreqFile SpectFile RecFile BValFile N1 N2 T OutFile where GridFile is a user defined grid input file defined in section 3 4 6 Examples The example input directory provides example input files For the following commands we assume that executable SweepFinder2 is located in the same directory as the example files There are three sets of files a background frequency spectrum from neutral simulations files generated from simulations with only background selection and files generated from simulations with both background and positive selection List of files from neutral simulations Neutral background sfs invar0 Derived SFS for counts x 1 2 n 1 Neutral background sfs invarl Derived SFS for counts x 1 2 n Neutral background sfs invar2 Derived SFS for counts x 0 1 n List of files from simulations with only background selection BGS noSweep SF 65 invar0 Allele frequencies for counts x 1 2 n 1 BGS noSweep SF 65 invarl Allele frequencies for counts x 1 2 n BGS noSweep SF 65 invar2 Allele freque
Download Pdf Manuals
Related Search