The German prosecutor working on the abduction of Madeleine McCann said the girl was killed by the main suspect by now, known as Christian B. Taking into consideration that he also said that they do not have the body, or even forensic evidence about the killing, I guess that the "non-forensic evidence" they have are testimonies of people very close to Christian B, especially in April and May, 2007. Protected witnesses? Anyway, it is not enough to charge the suspect.
Maybe the German prosecutor is considering the following: the day after the abduction, the Portuguese police collected samples from Madeleine's apartment, mainly hair. Here I list the human hair evidence (between brackets those which were used to obtain DNA profiles).
Zone 1. Madeleine's bedroom entrance, 32 hair (23).
Zone 2. Floor of Madeleine's bedside, 28 hair (10).
Zone 3. On Madeleine's bed, 4 hair (1).
Zone 4. Floor beside the bed next to the window, 15 hair (14).
Zone 6. Floor of the living room, 31 hair (22).
Zone 7. Entrance hallway to apartment, 58 hair (42).
Zone 2. Floor of Madeleine's bedside, 28 hair (10).
Zone 3. On Madeleine's bed, 4 hair (1).
Zone 4. Floor beside the bed next to the window, 15 hair (14).
Zone 6. Floor of the living room, 31 hair (22).
Zone 7. Entrance hallway to apartment, 58 hair (42).
From the 112 profiles, only 1 was nuclear DNA (nuDNA) and corresponded to Madeleine's father. The rest 111 profiles were mitochondrial DNA (mtDNA). Unlike the nuDNA, the mtDNA has almost a null value by itself to identify a person.
The reason why there are 111 of mtDNA profiles, and only one nuDNA profile is that, in 2007, was not posible to extract nuDNA form hair shaft without follicle. But in 2020, it is posible to obtain single-nucleotide polymorphism (SNP) nuDNA profiles from hair shaft without follicle.
Besides, another type of DNA analysis, the Short Tandem Repeat (STR) nuDNA one is not suitable because it uses repetitions of nucleotides for each marker, and nuDNA are fragmented, degraded, in hair shafts (while SNP uses only one nucleotide for each marker).
Besides, another type of DNA analysis, the Short Tandem Repeat (STR) nuDNA one is not suitable because it uses repetitions of nucleotides for each marker, and nuDNA are fragmented, degraded, in hair shafts (while SNP uses only one nucleotide for each marker).
DNA databases are implemented using STR analysis profiles, but if there is a suspect, it is posible to apply SNP analysis to both the evidence and the suspect. The question is: have the police some of those 167 hair (or part of them) available, or all were destroyed while performing the first analysis?