The microbiota play a vital role in the organism’s survival: the disruption of the equilibrium between the host and its microbiota can lead to the development of diseases, such as cancer. Nonetheless, it remains uncertain whether specific patterns of microbial colonisation are associated with the characteristics of tumours. This study introduces an approach for estimating microbial signals within human Next Generation sequencing (NGS) data, to clarify the link between microbes and the properties of tumours. Human NGS data may be susceptible to contamination and technical issues, so we have implemented controls to identify the most appropriate approaches to understand the overall microbial composition trends. We conducted the analyses on TCGA data and further evaluated it using a cohort of colon cancer patients. The majority of the identified microbes were bacteria, while only a small proportion of the signals could be attributed to viruses, eukaryota, or archaea. Leveraging the higher detection of bacterial signals, we were able to establish associations between bacterial compositions and several tumour properties, including survival rates, tumour location, microsatellite instability and consensus molecular subtype in colon tumours. Moreover, our findings suggest potential mechanisms through which colon tumours are linked to bacteria, such as interactions with immune cells and bacterial pathways. However, only a limited number of modest associations were identified in other cancer types. Through the concurrent analysis of tumour properties along with the composition of the microbiome, we enabled the exploration of the relationship between microbiota and tumours with a methodology that can improve patient stratification.
RECONSTRUCTION OF THE CONDITION- AND LOCATION-SPECIFIC COLON CANCER MICROBIOME FROM HUMAN RNA SEQUENCING DATA / G. Sambruni ; tutor: M. Schaefer; co-tutor: PG. Pelicci ; phd coordinator: S. Minucci. Dipartimento di Oncologia ed Emato-Oncologia, 2023. 34. ciclo, Anno Accademico 2022.
RECONSTRUCTION OF THE CONDITION- AND LOCATION-SPECIFIC COLON CANCER MICROBIOME FROM HUMAN RNA SEQUENCING DATA
G. Sambruni
2023
Abstract
The microbiota play a vital role in the organism’s survival: the disruption of the equilibrium between the host and its microbiota can lead to the development of diseases, such as cancer. Nonetheless, it remains uncertain whether specific patterns of microbial colonisation are associated with the characteristics of tumours. This study introduces an approach for estimating microbial signals within human Next Generation sequencing (NGS) data, to clarify the link between microbes and the properties of tumours. Human NGS data may be susceptible to contamination and technical issues, so we have implemented controls to identify the most appropriate approaches to understand the overall microbial composition trends. We conducted the analyses on TCGA data and further evaluated it using a cohort of colon cancer patients. The majority of the identified microbes were bacteria, while only a small proportion of the signals could be attributed to viruses, eukaryota, or archaea. Leveraging the higher detection of bacterial signals, we were able to establish associations between bacterial compositions and several tumour properties, including survival rates, tumour location, microsatellite instability and consensus molecular subtype in colon tumours. Moreover, our findings suggest potential mechanisms through which colon tumours are linked to bacteria, such as interactions with immune cells and bacterial pathways. However, only a limited number of modest associations were identified in other cancer types. Through the concurrent analysis of tumour properties along with the composition of the microbiome, we enabled the exploration of the relationship between microbiota and tumours with a methodology that can improve patient stratification.File | Dimensione | Formato | |
---|---|---|---|
phd_unimi_R12420.pdf
accesso aperto
Tipologia:
Altro
Dimensione
10.29 MB
Formato
Adobe PDF
|
10.29 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.