t.vandeuren@maastrichtuniversity.nl
26 november 2020 bij 16:42- Antwoorden
Hi Mirella, nice presentation. How do you think the decrease in actin skeleton expression remodelling will affect diabetes patients? And is an increase in interferon-y necessarily negative or could it also protect against viral infections?
M. Kalafati (Maastricht)
26 november 2020 bij 17:02- Antwoorden
Thank you for your question! Remodelling of actin cytoskeleton is a characteristic of adipose tissue dysfunction which has been previously associated with insulin resistant or diabetic individuals. This usually relates to a worsened metabolic profile. Since our data is in blood we hypothesize that downregulation of the actin cytoskeleton reflects adipose tissue dysfunction.
The interferon-related signature is not necessarily a negative outcome, it shows that these individuals are characterised by this profile. Since these data are in blood, we can hypothesize that these individuals are characterised by increased systemic inflammation, maybe due to an immune response. Obese individuals are prone to infections and have a worsened immune response to viral and bacterial infections,. Therefore, this signature may provide targets for more personalized risk classification.
Kristiaan Wouters
26 november 2020 bij 17:36- Antwoorden
Dear Mirella, very interesting approach. I have a few questions. There seems to be a relatively low amount of neutrophils via your calculations. Did you validate the calculations in any way?
Is it also possible to look at monocyte subsets?
How exactly should I interpret the correction for WBCs? Is there any way of estimating which immnune cells contribute most likely to the transcription signatures you found?
M. Kalafati (Maastricht)
26 november 2020 bij 17:59- Antwoorden
Thank you for your questions Kristiaan!
To answer to your first question, I haven’t not looked into more detail in regards to the low neutrophil count mainly because it was not significantly different between the insulin resistant and insulin sensitive individuals. Therefore, I have not validated this outcome. That is a good point though. What would you expect as a number to characterise these individuals?
It is important to mention here that I have checked in literature for the monocytes and that aligns with what it is reported. Monocytosis as well is associated with insulin resistance. I cannot look at monocyte subset with this approach unfortunately as this information is not provided by the algorithm.
The correction for WBC you can interpret as follows, when the effects of cell type proportional changes are eliminated, the changes in gene expression that remains are more likely to represent altered levels of gene transcription within the cells. Therefore, when we adjust for this changes especially in a tissue such as blood we obtain information that is not entirely driven by the differences in cell proportions.
With the current data and information I cannot estimate which immune cells contribute most likely to this transcription pattern. There are ways to do this, for instance more research on how to efficiently perform single-sample deconvolution, where cell type-specific profiles are estimated for each individual sample rather than groups, as already proposed in cancer tissues or the development of algorithms capable of performing “deep deconvolution”, i.e. accurately estimating from a whole blood or PBMC sample, the proportions and expression patterns of a greater number of cell subsets, going further down into the hematopoietic tree (T-regs, naive, memory, and effector cell subsets).
Kristiaan Wouters
26 november 2020 bij 18:12- Antwoorden
Ok thank you.
In a similar dataset (maastricht Study) we get anywhere between 40% and 70% of neutrophils.
For potential analysis of cell-specific gene expression, would it perhaps make sense to first try it with known immune cell marker genes and see how they relate with the epigenetic signature?
Anyway, would be happy to brainstorm on this anytime.
Regards
M. Kalafati (Maastricht)
26 november 2020 bij 18:16- Antwoorden
I see. I will look into the numbers of neutrophils and what are the numbers reported in literature.
Your suggestion is very good indeed and I would be happy to discuss further! Thank you for your nice questions and suggestions!
Rinke.stienstra@wur.nl
26 november 2020 bij 18:13- Antwoorden
Very interesting presentation! What would be your hypothesis related to potential factors that would lead to differential regulation of ISGs? Is this directly caused by altered levels of insulin or glucose ?
M. Kalafati (Maastricht)
26 november 2020 bij 18:27- Antwoorden
Thank you for your question!It is difficult to say. Our hypothesis based on the literature is that this interferon signature suggests increased systemic inflammation (possibly via an immune response) and could reflect organ dysfunction (e.g. adipose tissue). Since this is a cross-sectional study, I have to be very cautious in directly attributing any of these results to altered levels of insulin or glucose.
8 Reacties
Hi Mirella, nice presentation. How do you think the decrease in actin skeleton expression remodelling will affect diabetes patients? And is an increase in interferon-y necessarily negative or could it also protect against viral infections?
Thank you for your question! Remodelling of actin cytoskeleton is a characteristic of adipose tissue dysfunction which has been previously associated with insulin resistant or diabetic individuals. This usually relates to a worsened metabolic profile. Since our data is in blood we hypothesize that downregulation of the actin cytoskeleton reflects adipose tissue dysfunction.
The interferon-related signature is not necessarily a negative outcome, it shows that these individuals are characterised by this profile. Since these data are in blood, we can hypothesize that these individuals are characterised by increased systemic inflammation, maybe due to an immune response. Obese individuals are prone to infections and have a worsened immune response to viral and bacterial infections,. Therefore, this signature may provide targets for more personalized risk classification.
Dear Mirella, very interesting approach. I have a few questions. There seems to be a relatively low amount of neutrophils via your calculations. Did you validate the calculations in any way?
Is it also possible to look at monocyte subsets?
How exactly should I interpret the correction for WBCs? Is there any way of estimating which immnune cells contribute most likely to the transcription signatures you found?
Thank you for your questions Kristiaan!
To answer to your first question, I haven’t not looked into more detail in regards to the low neutrophil count mainly because it was not significantly different between the insulin resistant and insulin sensitive individuals. Therefore, I have not validated this outcome. That is a good point though. What would you expect as a number to characterise these individuals?
It is important to mention here that I have checked in literature for the monocytes and that aligns with what it is reported. Monocytosis as well is associated with insulin resistance. I cannot look at monocyte subset with this approach unfortunately as this information is not provided by the algorithm.
The correction for WBC you can interpret as follows, when the effects of cell type proportional changes are eliminated, the changes in gene expression that remains are more likely to represent altered levels of gene transcription within the cells. Therefore, when we adjust for this changes especially in a tissue such as blood we obtain information that is not entirely driven by the differences in cell proportions.
With the current data and information I cannot estimate which immune cells contribute most likely to this transcription pattern. There are ways to do this, for instance more research on how to efficiently perform single-sample deconvolution, where cell type-specific profiles are estimated for each individual sample rather than groups, as already proposed in cancer tissues or the development of algorithms capable of performing “deep deconvolution”, i.e. accurately estimating from a whole blood or PBMC sample, the proportions and expression patterns of a greater number of cell subsets, going further down into the hematopoietic tree (T-regs, naive, memory, and effector cell subsets).
Ok thank you.
In a similar dataset (maastricht Study) we get anywhere between 40% and 70% of neutrophils.
For potential analysis of cell-specific gene expression, would it perhaps make sense to first try it with known immune cell marker genes and see how they relate with the epigenetic signature?
Anyway, would be happy to brainstorm on this anytime.
Regards
I see. I will look into the numbers of neutrophils and what are the numbers reported in literature.
Your suggestion is very good indeed and I would be happy to discuss further! Thank you for your nice questions and suggestions!
Very interesting presentation! What would be your hypothesis related to potential factors that would lead to differential regulation of ISGs? Is this directly caused by altered levels of insulin or glucose ?
Thank you for your question!It is difficult to say. Our hypothesis based on the literature is that this interferon signature suggests increased systemic inflammation (possibly via an immune response) and could reflect organ dysfunction (e.g. adipose tissue). Since this is a cross-sectional study, I have to be very cautious in directly attributing any of these results to altered levels of insulin or glucose.