## User: Andrew_McDavid

Andrew_McDavid •

**80**- Reputation:
**80**- Status:
- Trusted
- Location:
- Website:
- https://amcdavid.githu...
- Twitter:
- EquivMeasures
- Last seen:
- 4 days, 3 hours ago
- Joined:
- 1 year ago
- Email:
- A*************@urmc.rochester.edu

#### Posts by Andrew_McDavid

<prev
• 13 results •
page 1 of 2 •
next >

0

votes

1

answers

114

views

1

answers

... Can you be more specific when you say "interface for contrasts" The contrasts(colData(x)$strain) = "contr.sum" business is base R and determines the coding for your factors, hence interpretation of your coefficients. Venables & Ripley is the canonical reference for that.
Specifying linear co ...

written 9 days ago by
Andrew_McDavid •

**80**0

votes

1

answers

114

views

1

answers

...
You can fit such a model a "cellmean model" (zlm_cellmean = zlm(~ 0 + group, x)) though it is possibly a little less statistically efficient [1] than using contrasts that explicitly includes an intercept:
## assuming you manually crossed strain and condition to get group
zlm_dummy = zlm(~stra ...

written 10 days ago by
Andrew_McDavid •

**80**0

votes

1

answers

356

views

1

answers

... TL;DR
One of the classes you are comparing has no detectable expression for a gene. The log-fold change is trying to get an estimate of effect size, but that is not well-defined when we never observe expression in a class. P-values from lrTest or waldTest are still well-defined.
Fine print
From t ...

written 5 weeks ago by
Andrew_McDavid •

**80**0

votes

1

answer

356

views

1

answer

... I noticed that MAST sometimes has NA for the continuous coefficient. When this happens, it also has NA for the estimated fold change. I couldn't find much in the documentation about this, but I was wondering if someone can explain it?
This is a problem for me because I'm trying to select the most d ...

written 5 weeks ago by
Andrew_McDavid •

**80**0

votes

3

answers

274

views

3

answers

... You might also try
Bacher R, Chu L, Leng N, Gasch AP, Thomson JA, Stewart RM, Newton MA and Kendziorski C (2017). “SCnorm: robust normalization of single-cell RNA-seq data.” Nature Methods. https://www.nature.com/nmeth/journal/vaop/ncurrent/full/nmeth.4263.html.
Software is available in Bioconduct ...

written 6 weeks ago by
Andrew_McDavid •

**80**0

votes

1

answers

544

views

1

answers

... The model coefficients that are being tested for enrichment are the coefficients from the zlmfit objects. With A as a ref, then positive coefficients in the ZlmFit object means that the mean of B > mean of A. The GSEA is calculating the average ZlmFit coefficient inside the set. So a positive Z-s ...

written 8 weeks ago by
Andrew_McDavid •

**80**0

votes

1

answers

92

views

1

answers

... Our goal is to support any model that you can run with `lm`. Using the numeric values for your ordered factor corresponds to a very particular model: it will estimate the linear effect of dose. That is a restriction from the factor model, which allows the mean function to follow *any* effect of do ...

written 8 weeks ago by
Andrew_McDavid •

**80**0

votes

1

answers

92

views

1

answers

... First off, in base R, I have only dabbled with the ordered factor contrasts so don't know them well enough to give reliable advice about how to interpret them. That being said, there are two ways to apply them with MAST.
First, you can set the contrasts attribute to the column of interest:
librar ...

written 8 weeks ago by
Andrew_McDavid •

**80**0

votes

1

answers

141

views

1

answers

... SIMLR, and related methods that use the cell-to-cell distance matrix such as TSNE will capture non-linear structure of the data. Regressing out variables and using the deviance residuals (roughly speaking) makes each gene orthogonal to the nuisance covariates. But that does not make the distance mat ...

written 4 months ago by
Andrew_McDavid •

**80**0

votes

1

answer

141

views

1

answer

... I have dataset of 443 cells from 6 different healthy donors (136,92,117,44,28,26 cells from each donor respectively). We are trying to look at gene-gene correlations over the 443 cell samples. Now before looking at the correlations we clustered the cells to check if the inter cell variation is now l ...

written 4 months ago by
Andrew_McDavid •

**80**#### Latest awards to Andrew_McDavid

Scholar
5 months ago,
created an answer that has been accepted.
For A: MAST reported warning "Coefficients ... are never estimible and will be dropped.

Use of this site constitutes acceptance of our User
Agreement
and Privacy
Policy.

Powered by Biostar
version 2.2.0

Traffic: 105 users visited in the last hour