Advice on analysis software for Vectra multiIF (InForm vs. HALO)

[Abate Daga, Daniel]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear all,

I need advice on the use of softwares for the analysis of multiplexed IF images, obtained using the Vectra platform. I am new to digital image analysis, so I’ve relied on recommendations by my collaborators. We started processing the images using InForm (cell segmentation, etc), followed by data extraction using FCS Express Image, or similar. Halfway through the process, I was told that reviewers would not accept data analyzed with InForm for publication, and that I absolutely must use Halo instead.

Can anyone provide their views on this issue? Would you automatically reject data analyzed with InForm if you reviewed a manuscript for publication?

Thank you in advance.

Best,

Daniel


Daniel Abate-Daga, Ph.D.
Assistant Member, Immunology Program
H. Lee Moffitt Cancer Center and Research Institute
tel:  [phone]| fax:  [phone]| email: [hidden email]


Please consider the environment before printing this email.

This transmission may be confidential or protected from disclosure and is only for review and use by the intended recipient. Access by anyone else is unauthorized. Any unauthorized reader is hereby notified that any review, use, dissemination, disclosure or copying of this information, or any act or omission taken in reliance on it, is prohibited and may be unlawful. If you received this transmission in error, please notify the sender immediately. Thank you.

[Mike Nelson]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Daniel,

Was there a reason given for the rejection of data from InForm? I had not
heard of that before. I am a big fan of open source solutions, but HALO is
no more open than InForm. Was it specifically InForm "in general" that was
the problem or some aspect of the way InForm was being used in your case?

I am aware of at least one group here at City of Hope that both stitches
the individual fields of view back into a whole slide image and analyzes
the data from Vectra multiplex using QuPath and occasionally R scripts.
Some new releases like CytoMAP allow the use of tSNE/PHATE plots with data
aggregated from multiple images, on which you can gate the results and view
those gates back in QuPath. It is not a clean connection right now,
involving two export and import steps, but it is not incredibly complex
either. CytoMAP also has some nice cluster analysis options that can be
investigated using a heatmap for various markers.
https://forum.image.sc/t/cytomap-newbie-questions/42675/10
I have not yet written up a full description of the link and data transfer
between the two programs, as I am hoping for a cleaner way to get heatmaps
of the makeup of individual gates.

Cheers,
Mike

On Tue, Sep 15, 2020 at 10:21 AM Abate Daga, Daniel <
[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Dear all,
>
> I need advice on the use of softwares for the analysis of multiplexed IF
> images, obtained using the Vectra platform. I am new to digital image
> analysis, so I’ve relied on recommendations by my collaborators. We started
> processing the images using InForm (cell segmentation, etc), followed by
> data extraction using FCS Express Image, or similar. Halfway through the
> process, I was told that reviewers would not accept data analyzed with
> InForm for publication, and that I absolutely must use Halo instead.
>
> Can anyone provide their views on this issue? Would you automatically
> reject data analyzed with InForm if you reviewed a manuscript for
> publication?
>
> Thank you in advance.
>
> Best,
>
> Daniel
>
>
> Daniel Abate-Daga, Ph.D.
> Assistant Member, Immunology Program
> H. Lee Moffitt Cancer Center and Research Institute
> tel:  [phone]| fax:  [phone]| email:
> [hidden email]
>
>
> Please consider the environment before printing this email.
>
> This transmission may be confidential or protected from disclosure and is
> only for review and use by the intended recipient. Access by anyone else is
> unauthorized. Any unauthorized reader is hereby notified that any review,
> use, dissemination, disclosure or copying of this information, or any act
> or omission taken in reliance on it, is prohibited and may be unlawful. If
> you received this transmission in error, please notify the sender
> immediately. Thank you.
>

[Abate Daga, Daniel]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear Mike,

Thanks for your quick response. I didn't get any technical detail of why this person considered InForm inappropriate for the analysis. He stated that the software was not developed for that type of analysis and that the manuscript would be rejected by a reviewer versed in image analysis. Your response is in line with others that I've got. Thanks so much for your time, and also for forwarding this question to your colleague at CoH.

Best,

Daniel

[Sripad Ram-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

 Hi Daniel,
I have a fair share of experience using inForm, HALO and QuPath for
multiplex image analysis. As Mike already mentioned, both packages are
proprietary and you may be better off using QuPath although it is not
always easy to figure things out in that software (but also not too hard).

Can you elaborate on what you are trying to accomplish? Are you trying to
phenotype and count cells or are you also interested in spatial analysis?
You can generate a lot of numbers and some of which might even show
statistically significant differences (if you are comparing, say, control
and treated groups). However, you need to be careful as to how you analyze
your data.

I am not aware of any reason why inForm data cannot (or should not) be used
for publications. There are lots of papers that have reported image
analysis results from inForm.

In my experience, both inForm and HALO will do a comparable of job
segmenting nuclei and detecting membrane. Inform has a more sophisticated
phenotype classifier (which is surprisingly open source --- uses the RLMNet
package in R) but it needs to be properly configured and adequately
trained. In this regard, HALO might be better since it uses hard thresholds
to phenotype cells. While intensity based thresholding has its limitations,
it is easy to implement and relatively straightforward to verify.

Just my two cents.

Hope this helps.

Sripad




I gather that you are doing some multiplex image analysis.  Can you briefly
describe what you are trying to  accomplish?

Thanks.

Sripad


On Tue, Sep 15, 2020 at 10:52 AM Daniel Abate-Daga <
[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Dear Mike,
>
> Thanks for your quick response. I didn't get any technical detail of why
> this person considered InForm inappropriate for the analysis. He stated
> that the software was not developed for that type of analysis and that the
> manuscript would be rejected by a reviewer versed in image analysis. Your
> response is in line with others that I've got. Thanks so much for your
> time, and also for forwarding this question to your colleague at CoH.
>
> Best,
>
> Daniel
>

[Abate Daga, Daniel]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Sripad,

Thanks for your input. The intended use is for cell segmentation and phenotyping.

Best,

Daniel

[Tam Hong Nguyen]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Daniel,

Just wanted to add our experience in this area to the mix. We have guided several of our researchers through using  Inform, FCS Express, QuPath and HALO for analyzing their multiplexed tissue samples. As with previous comments I have not heard of any instances of Journals not accepting data processed/analysed with Inform Software. If you have acquired you multispectral images on the Vectra then you will need to use Inform to perform unmxing and autofluorescence removal as a first step anyway. From there you could continue with tissue and cell segmentation and machine learning-based phenotyping (classification of cells based on their marker fluorescence profile).

Alternatively, after doing unmixing you could export the images out and use third party software, such as QuPath or HALO, to do Tissue/Cell segmentation/phenotyping. These processing steps would very similar to that in InForm, so again I not sure why Inform would be singled out as unacceptable approach by reviewers.
My current preference is to use QuPath due to its segmentation ability, flexibility and being open source.

When we first got the Vectra platform the vendors provided Spotfire as a software for visualizing data that was coming out of InForm. However, our users found it to be too clunky and cumbersome so, like you, we migrated to FCS Express Plus/Image, which incidently we worked closely with Denovo to develop the plugin to load and view Vectra/Inform images and data.

Hope this helps to answer some of your questions and happy to chat in more detail if you wish.

Regards

Tam

Tam Nguyen  | Ph.D
Senior Microscopy Analyst | Flow Cytometry and Imaging Facility
QIMR Berghofer Medical Research Institute
 
t [phone] | m + 0421409282 | f + 61 7 3362 0107
e [hidden email]  |  www.qimr.edu.au
300 Herston Road, Herston QLD 4006
Locked Bag 2000 Royal Brisbane Hospital, QLD 4029
 




-----Original Message-----
From: Confocal Microscopy List <[hidden email]> On Behalf Of Abate Daga, Daniel
Sent: Wednesday, September 16, 2020 3:00 AM
To: [hidden email]
Subject: Advice on analysis software for Vectra multiIF (InForm vs. HALO)

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear all,

I need advice on the use of softwares for the analysis of multiplexed IF images, obtained using the Vectra platform. I am new to digital image analysis, so I’ve relied on recommendations by my collaborators. We started processing the images using InForm (cell segmentation, etc), followed by data extraction using FCS Express Image, or similar. Halfway through the process, I was told that reviewers would not accept data analyzed with InForm for publication, and that I absolutely must use Halo instead.

Can anyone provide their views on this issue? Would you automatically reject data analyzed with InForm if you reviewed a manuscript for publication?

Thank you in advance.

Best,

Daniel


Daniel Abate-Daga, Ph.D.
Assistant Member, Immunology Program
H. Lee Moffitt Cancer Center and Research Institute
tel:  [phone]| fax:  [phone]| email: [hidden email]


Please consider the environment before printing this email.

This transmission may be confidential or protected from disclosure and is only for review and use by the intended recipient. Access by anyone else is unauthorized. Any unauthorized reader is hereby notified that any review, use, dissemination, disclosure or copying of this information, or any act or omission taken in reliance on it, is prohibited and may be unlawful. If you received this transmission in error, please notify the sender immediately. Thank you.
[EXTERNAL EMAIL] This message originates from an external email address, please exercise caution when clicking any links or opening attachments. If you believe the sender is impersonating someone at QIMR Berghofer, please forward this message to [hidden email].

[Valeria Mezzano]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Daniel,
I'll add our experience to the mix!
We analyze data from Vectra 3.0 and Vectra-Polaris at our core constantly.
As you know InForm must be used to obtain written component_data.tif files because neither vectra 3.0 NOR Vectra-Polaris output spectrally-unmixed image files. Therefore that first step must be performed in InForm.

Further analysis (image segmentation and classification) can then proceed in InForm like you are doing and looking at the segmented data in FCSExpress or the component_data.tiff file can be analyzed with any other image analysis software of your choice.

I am shocked to hear that anyone would categorically disregard one software versus another for this purpose. We have analyzed, and published, vectra data with InForm (all the way to phenotyping phase), Visiopharm (using only component_data.tifs) and Fiji/ImageJ2. Needless to say that the output regardless of software must be thoroughly quality controlled and the steps involved should be disclosed.
 
I would be very interested to hear if any other people in this list have heard the same comment and the reason for it.

If you would like further discussion please email me at [hidden email]


Best!

Valeria

Experimental Pathology, Div. Advanced Research Technologies,
NYU LMC

[Zalavadia, Ajaykumar]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Daniel, I agree with Tam with regards to QuPath. I have used QuPath and Halo. In the multiplex tissue scans I have worked with, QuPath + StarDist combination was more accurate at finding and segmenting nuclei using DAPI channel. If nuclei are well separated and evenly stained QuPath with StarDist may or may not give you that benefit compare to HALO. I am sure HALO or InForm would do similar analysis (phenotyping) without an issue and are as much valid. One thing I would like to point out is that it may get difficult to keep track of the quality of output when you are dealing with 300 to 400 individual fields/images per section in InForm (In my case it was around 10000 fields from 25 slides) compare to HALO or QuPath where you can quickly visualize whole tissue section with phenotype results overlaid.

My current preference is also QuPath, due to being open source, flexibility in exporting data, and ability to write custom scripts. Also, Pete Bankhead (QuPath developer) and few other members are very active and helpful at image.sc forums.

Regards,
Ajay

Ajay Zalavadia, Ph.D.
Imaging Specialist, Imaging Core  |  Lerner Research Institute
9500 Euclid Avenue NB10  |  Cleveland, OH 44195
Phone: [phone] | email: [hidden email]<mailto:[hidden email]>


From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Valeria Mezzano
Sent: Wednesday, September 16, 2020 12:31 PM
To: [hidden email]
Subject: [EXT] Re: Advice on analysis software for Vectra multiIF (InForm vs. HALO)

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy<http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy>
Post images on http://www.imgur.com<http://www.imgur.com> and include the link in your posting.
*****

Hi Daniel,
I'll add our experience to the mix!
We analyze data from Vectra 3.0 and Vectra-Polaris at our core constantly.
As you know InForm must be used to obtain written component_data.tif files because neither vectra 3.0 NOR Vectra-Polaris output spectrally-unmixed image files. Therefore that first step must be performed in InForm.

Further analysis (image segmentation and classification) can then proceed in InForm like you are doing and looking at the segmented data in FCSExpress or the component_data.tiff file can be analyzed with any other image analysis software of your choice.

I am shocked to hear that anyone would categorically disregard one software versus another for this purpose. We have analyzed, and published, vectra data with InForm (all the way to phenotyping phase), Visiopharm (using only component_data.tifs) and Fiji/ImageJ2. Needless to say that the output regardless of software must be thoroughly quality controlled and the steps involved should be disclosed.

I would be very interested to hear if any other people in this list have heard the same comment and the reason for it.

If you would like further discussion please email me at [hidden email]<mailto:[hidden email]>


Best!

Valeria

Experimental Pathology, Div. Advanced Research Technologies,
NYU LMC


Please consider the environment before printing this e-mail

Cleveland Clinic is currently ranked as the No. 2 hospital in the country by U.S. News & World Report (2017-2018). Visit us online at http://www.clevelandclinic.org for a complete listing of our services, staff and locations. Confidentiality Note: This message is intended for use only by the individual or entity to which it is addressed and may contain information that is privileged, confidential, and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient or the employee or agent responsible for delivering the message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy. Thank you.