Editing Final Report/Thesis 2019 (section)

==Task 5: Investigation on DNA matches==
===Aims===
The aims of this task is to investigate what results if the DNA match services provide on GEDmatch are conducted on high quality DNA kits, and how the degradation could effect the match results.

===Methods===
In task 2, the project has conducted DNA match examination on Somerton Man's DNA kit with multiple methods, but there is no match results for the his DNA reference file. In this task, one-to-many tool will be used again on 2 sample DNA files the project ordered from 23andMe, and the match results shall be recorded.
Then the DNA match tests would be conducted on the degraded files created in task 3. The top 30 match results for each degraded DNA kit would be recorded and compare with the results of their original kit. A false positives and false negatives test would be conducted to show the change of match results during the degradation process. In this case, false positives would be match kits that are in the degraded kit's match results but not in original kit's result. And false negatives would present kits that are matched with original kit but not with the degraded one. An example is presented for a clear understanding. There are 5 kits A, B, C, D and E matched with the original kit, and kits A, B, C, M and N are matched with a degraded kit. Then the false positives for this degraded kit are D and E, and the false negatives are kits M and N. A line graphs of the number of false positives and false negatives against the percentage of SNPs removed will be created to show how degradation process effect the match results.

===Results and discussion===
Both DNA samples were successfully found their matched DNA kits in the database. Sample 1 have 8182 match kits and there are 5968 DNA files are found related to the sample 2. Top 30 match kits of sample 1 are shown in figure 20. The column Kit, Name and Email indicate the kit number, name of the kit and email of kit's owner. Column Total cM shows the total centimorgan which is a measure of genetic linkage between the 2 DNA kits. Note that the top 30 match kits are the kits with largest total centimorgan. Last but not least, the Overlap column present how many SNPs were used in the comparison between 2 kits. 
[[File:match_results_sample1.png|thumb|600px|center|Figure 20: match results of DNA sample 1]]
Next, the top 30 match kits for each degraded DNA reference files are recorded and the false negatives and false positives are calculated. Since all degraded files except the degraded files with 10% SNPs remaining have more than 30 match kits, the number of false negatives and false positives are same. The degraded files with 10% SNPs remaining have no match results. There are 4 degradation strategies introduced in task 3, therefore 4 sets of false negatives and false positives are provided for analysis. Figure 21 present line graph of the number of false negatives and false positives against degradation levels. The number of false negatives and false positives are the mean of 4 sets of data. Degradation level of 10% SNP remaining is not involved in the graph due to 0 match result. Similar graph which was done by last year's project were shown in figure 22. The DNA sample used in figure 22 is a completely different one from the samples used in figure 21. According to both graphs, the number of false positives and false negatives for different DNA samples are not same. But the trend are similar. As more SNPs are removed, the amount of false positives and false negatives increases until 50% SNPs are removed. When there is more than half amount of SNPs being removed, the number of false positives and false negatives reaches maximum of 30 which indicate that the match results of original kits and degraded kits are totally different at these levels. These results show that as more SNPs removed from the original DNA reference file, the match results would be more inaccuracy. And when there is only half amount of SNPs remaining in the DNA kit, the match results would be totally different and be unreliable. Moreover, when 10% of SNPs are removed, more than half of match results would be different which indicates that even a small amount of SNPs being removed could result a huge difference in DNA match test.
[[File:false1.png|thumb|600px|center|Figure 21: False Positives and False Negative test for DNA sample 1 and 2]]
[[File:false2.png|thumb|600px|center|Figure 22: False Positives and False Negative test from previous year [14]]]

===Conclusion===
According to the findings in this task, the project can conclude that it would require a high level quality of DNA which would be at least more than 90% SNPs are available in the DNA reference file to receive a reliable DNA match results. Only a small amount of SNPs in the DNA file are changed could result a significant affect on DNA match results. In another case, if the Somerton Man's DNA reference file is available to be recovered to more than 20% SNPs remaining, there could be DNA kits found related to him. And If the Somerton Man's DNA kit could be recoverd to a level of 60% SNPs remaining, part of his DNA match results can be reliable.