CDBN Individual Heterozygousity

Just quickly exploring two questions: How much heterozygousity is there in the bean samples, and how much does heterozygousity vary by bean variety?

I used Tassel to obtain a summary table of heterozygousity for each “taxa” (i.e., individual) in the vcf file. This vcf file was filtered so that each SNP present was genotyped in >33% of individuals. It was also filtered so that SNPs had to have a MAF of at least 1%. I don’t think either of these filters should change the analysis significantly… though these filters do remove about 2.5 million SNPs from the analysis.

This analysis is genome-wide, on 639127 SNPs. Here’s a basic summary of the data file. The proportion of heterozygousity in each individual varies from almost nothing to about 22%:

      Taxa                Taxa.Name   Number.of.Sites  Gametes.Missing   Proportion.Missing Number.Heterozygous
 Min.   :  0.0   CDBN_001_1543 :  1   Min.   :639127   Min.   :  36766   Min.   :0.02876    Min.   :     4     
 1st Qu.:150.8   CDBN_002_2242 :  1   1st Qu.:639127   1st Qu.: 207874   1st Qu.:0.16262    1st Qu.: 17351     
 Median :301.5   CDBN_003_5229 :  1   Median :639127   Median : 308012   Median :0.24096    Median : 31787     
 Mean   :301.5   CDBN_004_5501 :  1   Mean   :639127   Mean   : 356238   Mean   :0.27869    Mean   : 38999     
 3rd Qu.:452.2   CDBN_005_54028:  1   3rd Qu.:639127   3rd Qu.: 426188   3rd Qu.:0.33341    3rd Qu.: 55641     
 Max.   :603.0   CDBN_006_55012:  1   Max.   :639127   Max.   :1267156   Max.   :0.99132    Max.   :134793     
                 (Other)       :598                                                                            
 Proportion.Heterozygous            Inbreeding.Coefficient Inbreeding.Coefficient.Scaled.by.Missing
 Min.   :0.0002762       Inbreeding Coefficient:604        ICSBM:604                               
 1st Qu.:0.0418125                                                                                 
 Median :0.0659900                                                                                 
 Mean   :0.0759546                                                                                 
 3rd Qu.:0.1030775                                                                                 
 Max.   :0.2262500                                                                                 
                                                                                                   

Here’s a sorted table for the proportion of heterozygous sites per taxa.

(SNPs.noncom <- SNPs.604 %>%
  filter(!grepl('combined', Taxa.Name)) %>% # Selects 523 rows that were not combined data.
   select(Taxa.Name, Proportion.Missing, Proportion.Heterozygous) %>% 
   arrange(desc(Proportion.Heterozygous))        # Arrange the table so that the greatest heterozygousity comes first.
 )

The immediate problem is that there is a negative relationship between the proportion of sites missing and the proportion of heterozygous sites.

  ggplot(SNPs.noncom, mapping = aes(x = Proportion.Heterozygous, y = Proportion.Missing)) + 
     geom_point() + geom_smooth()

If we get rid of taxa with different percentages of sites that are missing, it looks like heterozygousity is quite high in these varieties - at least 7% heterozygousity on average, but probably upwards of 10% if we account for missing data or low coverage in some taxa.

(Het.summary <- SNPs.604 %>%
  filter(!grepl('combined', Taxa.Name)) %>%           # Selects 523 rows that were not combined data.
   arrange(desc(Proportion.Heterozygous)) %>%        # Arrange the table so that the greatest heterozygousity comes first.
   summarise(het.75percent.missing = mean(Proportion.Heterozygous[Proportion.Missing <= 0.75]),
             het.50percent.missing = mean(Proportion.Heterozygous[Proportion.Missing <= 0.50]),
             het.25percent.missing = mean(Proportion.Heterozygous[Proportion.Missing <= 0.25]),
             het.12.5percent.missing = mean(Proportion.Heterozygous[Proportion.Missing <= 0.125]),
             het.6.25percent.missing = mean(Proportion.Heterozygous[Proportion.Missing <= 0.0625]),
             no.6.25percent.missing = sum(!is.na(Proportion.Missing[Proportion.Missing <= 0.0625])))
   )

Ok, what about across positions in the genome? Are there any blocks in the genome that (for all taxa) are particularly heterozygous?

SNPs.sites <- read.table("C:/Users/ahm543/OneDrive/NSF Common Bean/CDBN Genomics/CDBN SNPs/8_Tassel/604_individual_genoSummary3.txt", sep = "\t", header = TRUE)
summary(SNP.sites)

Error in summary(SNP.sites) : object 'SNP.sites' not found

Here’s a summary of the genotypic data, averaged by site. The only really helpful thing here is that the Proportion.Heterozygous has really quite a low third quartile (0.09), which probably indicates there aren’t a huge number of problematic SNPs.

summary(SNPs.sites)

  Site.Number            Site.Name        Chromosome  Physical.Position  Number.of.Taxa Ref        Alt       
 Min.   :     0   S01_10007272:     1   Min.   :1.0   Min.   :     195   Min.   :604    A:130491   A:192829  
 1st Qu.:159782   S01_10007276:     1   1st Qu.:2.0   1st Qu.:11273856   1st Qu.:604    C:189176   C:126833  
 Median :319563   S01_10007277:     1   Median :4.0   Median :19925246   Median :604    G:188870   G:127192  
 Mean   :319563   S01_10007281:     1   Mean   :3.5   Mean   :20747115   Mean   :604    T:130590   T:192273  
 3rd Qu.:479345   S01_10007282:     1   3rd Qu.:5.0   3rd Qu.:28786009   3rd Qu.:604                         
 Max.   :639126   S01_10007291:     1   Max.   :7.0   Max.   :53433183   Max.   :604                         
                  (Other)     :639121                                                                        
 Major.Allele Major.Allele.Gametes Major.Allele.Proportion Major.Allele.Frequency Minor.Allele Minor.Allele.Gametes
 A:126368     Min.   : 191.0       Min.   :0.1581          Min.   :0.3031         A:197006     Min.   :  4.0       
 C:193706     1st Qu.: 580.0       1st Qu.:0.4801          1st Qu.:0.7125         C:122293     1st Qu.: 42.0       
 G:193374     Median : 692.0       Median :0.5728          Median :0.8475         G:122669     Median :132.0       
 T:125679     Mean   : 711.3       Mean   :0.5889          Mean   :0.8190         T:197159     Mean   :158.8       
              3rd Qu.: 839.0       3rd Qu.:0.6945          3rd Qu.:0.9503                      3rd Qu.:254.0       
              Max.   :1190.0       Max.   :0.9851          Max.   :0.9900                      Max.   :602.0       
                                                                                                                   
 Minor.Allele.Proportion Minor.Allele.Frequency Allele.3      Allele.3.Gametes  Allele.3.Proportion
 Min.   :0.00331         Min.   :0.01000        A   :  6323   Min.   :  0.000   Min.   :0.000000   
 1st Qu.:0.03477         1st Qu.:0.04917        C   :  3620   1st Qu.:  0.000   1st Qu.:0.000000   
 Median :0.10927         Median :0.15116        G   :  3870   Median :  0.000   Median :0.000000   
 Mean   :0.13143         Mean   :0.17963        T   :  6435   Mean   :  1.216   Mean   :0.001007   
 3rd Qu.:0.21026         3rd Qu.:0.28615        NA's:618879   3rd Qu.:  0.000   3rd Qu.:0.000000   
 Max.   :0.49834         Max.   :0.50000                      Max.   :356.000   Max.   :0.294700   
                                                                                                   
 Allele.3.Frequency Allele.4      Allele.4.Gametes    Allele.4.Proportion Allele.4.Frequency  Allele.5      
 Min.   :0.000000   A   :   210   Min.   :  0.00000   Min.   :0.000e+00   Min.   :0.000e+00   Mode:logical  
 1st Qu.:0.000000   C   :   180   1st Qu.:  0.00000   1st Qu.:0.000e+00   1st Qu.:0.000e+00   NA's:639127   
 Median :0.000000   G   :   178   Median :  0.00000   Median :0.000e+00   Median :0.000e+00                 
 Mean   :0.001313   T   :   230   Mean   :  0.02111   Mean   :1.748e-05   Mean   :2.149e-05                 
 3rd Qu.:0.000000   NA's:638329   3rd Qu.:  0.00000   3rd Qu.:0.000e+00   3rd Qu.:0.000e+00                 
 Max.   :0.330860                 Max.   :161.00000   Max.   :1.333e-01   Max.   :1.952e-01                 
                                                                                                            
 Allele.5.Gametes Allele.5.Proportion Allele.5.Frequency Allele.6       Allele.6.Gametes Allele.6.Proportion
 Min.   :0        Min.   :0           Min.   :0          Mode:logical   Min.   :0        Min.   :0          
 1st Qu.:0        1st Qu.:0           1st Qu.:0          NA's:639127    1st Qu.:0        1st Qu.:0          
 Median :0        Median :0           Median :0                         Median :0        Median :0          
 Mean   :0        Mean   :0           Mean   :0                         Mean   :0        Mean   :0          
 3rd Qu.:0        3rd Qu.:0           3rd Qu.:0                         3rd Qu.:0        3rd Qu.:0          
 Max.   :0        Max.   :0           Max.   :0                         Max.   :0        Max.   :0          
                                                                                                            
 Allele.6.Frequency Gametes.Missing Proportion.Missing Number.Heterozygous Proportion.Heterozygous
 Min.   :0          Min.   :  0.0   Min.   :0.0000     Min.   :  0.00      Min.   :0.00000        
 1st Qu.:0          1st Qu.:210.0   1st Qu.:0.1738     1st Qu.:  1.00      1st Qu.:0.00271        
 Median :0          Median :318.0   Median :0.2632     Median : 11.00      Median :0.02734        
 Mean   :0          Mean   :336.7   Mean   :0.2787     Mean   : 36.86      Mean   :0.07548        
 3rd Qu.:0          3rd Qu.:458.0   3rd Qu.:0.3791     3rd Qu.: 45.00      3rd Qu.:0.09978        
 Max.   :0          Max.   :808.0   Max.   :0.6689     Max.   :586.00      Max.   :0.97830        
                                                                                                  
 Inbreeding.Coefficient Inbreeding.Coefficient.Scaled.by.Missing
 TBD:639127             TBD:639127                              
                                                                
                                                                
                                                                
                                                                
                                                                
                                                                

If you plot heterozygousity by chromosome, you see that my f***ing SNP calling did not actually finish the way I thought it did! It only got to chromosome 7! Ok, I may have to request a >>48hr job to finish calling these bastards.

If you just look at the SNPs I have, you see there are some spikes in heterozygousity in a few places that we should be concerned about (end of Chr1, 3, and 5 - I bet you anything these are R-gene clusters!).

ggplot(data = SNPs.sites, mapping = aes(x = Physical.Position, y = Proportion.Heterozygous)) + 
  geom_hex() + geom_smooth(color = "white") +
    facet_wrap( ~ Chromosome)

Let’s zoom in on different parts of Chromosome 1 to see if these are really hotspots. They are, pretty convincingly.

Chr1 <- filter(SNPs.sites, Chromosome == 1)
Chr1.pt1 <- filter(Chr1, Physical.Position < 2000000)
Chr1.pt2 <- filter(Chr1, Physical.Position %in% c(2000000:4000000))
Chr1.pt3 <- filter(Chr1, Physical.Position %in% c(4000000:5200000))
ggplot(data = Chr1.pt1, mapping = aes(x = Physical.Position, y = Proportion.Heterozygous)) + 
  geom_hex(bins = 200)

ggplot(data = Chr1.pt2, mapping = aes(x = Physical.Position, y = Proportion.Heterozygous)) + 
  geom_hex(bins = 200)

ggplot(data = Chr1.pt3, mapping = aes(x = Physical.Position, y = Proportion.Heterozygous)) + 
  geom_hex(bins = 120)

Do these ‘heterozygousity hotspots’ map to frequency of major or minor alleles, or the proportion of missing alleles? There seems to perhaps be a relationship with all three. This makes me wonder if this pattern isn’t a heterozygousity problem so much as it is a reduced representation problem. What do you think??

ggplot(data = Chr1.pt1, mapping = aes(x = Physical.Position, y = Proportion.Heterozygous)) + 
     geom_hex(mapping = aes(x = Physical.Position, y = Major.Allele.Frequency), bins = 120) + geom_line(color = "red")

ggplot(data = Chr1.pt1, mapping = aes(x = Physical.Position, y = Proportion.Heterozygous)) + 
     geom_hex(mapping = aes(x = Physical.Position, y = Minor.Allele.Frequency), bins = 120) + geom_line(color = "red")

ggplot(data = Chr1.pt1, mapping = aes(x = Physical.Position, y = Proportion.Heterozygous)) + 
     geom_hex(mapping = aes(x = Physical.Position, y = Proportion.Missing), bins = 120) + geom_line(color = "red")