Omics

Question 1

Schizophrenia Working Group of the Psychiatric Genomics Consortium in 2014

Answer 1

published an analysis of almost 37,000 pts and 113,000 controls investigating SNPs

They identified 108 loci which implicated around 600 genes with a statistically significant association to schizophrenia.

o 83 of these conservatively defined loci had not been implicated before, thus new associations were also found with CACNA1C and CACNB2, which code for voltage-gated calcium channel subunits

o Moreover, associations were also significantly enriched at enhancers active in cells with important immune functions, particularly B-lymphocytes (CD20 and CD19), suggesting that the immune system could somehow be involved in schizophrenia risk

Question 2

Lohoff et al. (2020)

Answer 2

recently conducted an EWAS with 625 patients with alcohol use disorder, and found a new association with the long non-coding RNA growth arrest specific five gene (GAS5), which has been implicated in regulating transcriptional activity of the glucocorticoid receptor and has roles related to cancer, apoptosis and immune function.

Question 3

Lundby et al. (2014)

Answer 3

used tissue-specific quantitative interaction proteomics together with GWAS data to map a network of genes involved in long QT syndrome.

• Firstly, they selected 5 genes that cause Mendelian forms of long QT syndrome, and immunoprecipitated the respective proteins from lysates of mouse hearts
• They then identified 584 proteins (many of which had previously been shown to interact with ion channels) that co-precipitated with these 5 target proteins using mass spectrometry, indicative of possible protein-protein interactions.
• They compared this collection of proteins with genes from 35 GWAS loci for common forms of QT-interval variation
• Overall, they found 12 genes that overlapped between the two data sets. This suggests a causative link in the locus, and provides more powerful data than what could be found using genomics alone.

Question 4

Kandoth et al. (2013)

Answer 4

in a landmark paper analysed data from The Cancer Genome Atlas (TCGA) for point mutations and small insertions or deletions, for 3,281 tumours across 12 tumour types.
o They identified 127 genes that were mutated at a higher frequency, and 93% of samples had one of these significantly mutated genes.
o TP53 was the most commonly mutated gene (42% of samples), and they identified 66 genes as ‘mut-driver genes’.

Question 5

Nature in February 2020 by The ICGC/TCGA Pan-Cancer Analysis of Whole Genomes Consortium

Answer 5

analysed 2,658 whole-cancer genomes and matching normal tissues across 38 tumour types (compared to 12 in the Kandoth paper).
o They identified that cancer genomes contained on average 4-5 driver mutations, but in about 5% of cases, no driver mutation was identified, highlighting a gap in our knowledge, that we do not understand every cancer-driving mutation

Question 6

Stephens et al. (2011)

Answer 6

used next-generation sequencing on 10 patients with chronic leukaemia. They sequenced both ends of 50-100 million genomic DNA fragments per sample, aligned these to a reference genome, and identified putative rearrangements.

• They identified one patient with 42 somatic genomic rearrangements, almost all geographically localised to focal points on a few chromosomes.
• By further analysis of SNP arrays of 746 cancer cell lines, they found that at least 2-3% of cancers had such a complex rearrangement in a single chromosome, and this was not isolated to a particular cancer type.
• Thus, is was very unlikely that these mutations occurred individually. Rather, a single catastrophic event could explain this clustering and the finding that the final configuration of the chromosome was restricted to two (rarely three) copy number states (sequential mutations would cause a greater number of breakpoints)