Posts Tagged ImageJ

Comparison Between SIV and HIV: Shalanda and Raphael

We want to show differences between HIV and SIV as demonstrated by data we present. Using ImageJ we counted the number of spikes from tomograms of HIV-1 and SIVmac239 published in PNAS (Zhu, P. et al, PNAS, 100(26): 15812, 2003). We then looked at available PDB data for viral spikes and present PyMOL and Consurf images of SIV gp41, HIV gp41 and HIV gp120 protein components. Both SIV and HIV have gp120 and gp41 which is cleaved from gp160. Our concept map created with Cmap uses current data on SIV and HIV in Proteopedia.

Tags: ,

Caroline Image Analysis

Image Analysis Blog

The ascospores of Neosartorya stramenia (image C) and Talaromyces flavus (image D, Eliades et al. 2006)  were measured (length at their widest point excluding protruding surface structures) using ImageJ with the given scale bar of 2 µm . These results show that N. stramenia ascospores (average diameter 2.16 µm) are smaller than T. flavus (average diameter 3.28 µm).


Protein visualization (Rachel / Thomas)

Solved crystal structure of BRCA1 BRCT (made in Jmol):

Electrostatic map of BRCA1 BRCT (made in Pymol):

Sequence of BRCA1 BRCT showing features such as beta strands and alpha helices:

The structure has a diameter of ~27 A.  ~55 A in length.

There are a variety of secondary structures including beta turns, beta strands and alpha helices.

BRCA1 is composed of 215 residues.

Conserf run (MSA):

Dark purple regions are conserved.  Notice ligands binding in that region.

Tags: , , , , , , ,


Image J is a useful tool but like all other tools it has its drawbacks; which was discovered when examining selected pictures. There are limitations on what can be counted and how it will be counted. Even with this shortcoming the program is fantastic if inside certain parameters.

The source of the image is:,r:3,s:420

The strategy used for analysis was counting



Inferences that can me made from the count could be the number of cancer cells in a blood smear. If this is indeed a smear. If it isn’t, the picture can still give the number of cells present by counting the nuclei present in the given space.


Counting Soldiers

I pulled a satellite image from Google maps of a section of Arlington National Cemetery:

By using the scale provided by Google and ImageJ, I determined that the average length between rows of graves is 9.7 feet, and the average length between headstones is 5.9 feet.

I then cropped the image to an area (~35,200 sqft) with no visual obstructions (trees, scale bar, etc.) in order to effectively count the headstones.

I performed 2 counts in this area. The first, I restricted the area of particle counted to the range 0.5-2.0 sqft. This was to estimate the number of headstone, and therefor the number of total graves in this area. The second count, I restricted the area of particles counted to larger than 3.0 sqft. This was to count the large light patches that in the image reflect recently moved earth, thereby giving an estimate of new graves in the area.

Total graves: 704

New graves: 3

Extrapolating this data out, it would estimate that there is 1 grave for every 50 sqft, and 1 new grave every 12,000 sqft.

According to Wikipedia, Arlington National Cemetery is 624 acres, or 27,900,000 sqft). So, if you further project my estimations (and assume that the same density is present across the entire cemetery), then there are 54,000 grave sites, 2300 of which are newly established.


Flamingo counting

The image I used for the ImageJ analysis was of a group of flamingos.

I converted the image to 8 bit and set the threshold at 2 different levels to attempt to get the best count.  An issue was that some of birds were very close together and were counted as one.  I did not exclude edges since there are birds on the edges, but ones that are mostly off the photo were not counted.  The count is going to be underestimating the actual number due to these factors.  I attempted to set the threshold to separate them as much as possible without losing any of the birds, used a pixel range of 5-infinity with circularity set to a minimum of 0 (since they are not round at all), and the best count was 304.


Slice Count Total Area Average Size Area Fraction Mean Major Minor Angle
flamingo.jpg 304 3535 11.628 4.3 254.958-
flamingo.jpg 297 3368 11.34 4.1 254.957-


Image J

The black and white picture separated most fish into 2 or 3 stripes.  As an estimate I took the results and divided by 2.  The result from Image J was 659 individual strips so I estimate about 330 fish.

Slice Count Total Area Average Size Area Fraction Mean
tropical fish.jpg 659 178569 270.97 9.3 254.447-

The image was downloaded from,%20Tahiti%20pictures%20underwater%20photos.php


The link below is my analysis of myeloma cells using image J.

Ann Wells PEER image J


Using ImageJ

I found this picture of Campylobacter jejuni taken with a scanning electron microscope. Using ImageJ, I measured the length of the seven whole cells, excluding the flagella. I found that 3 of the smaller cells all had lengths around 2.7μm, while the average length of all the cells was 4.0μm. Obviously the very long cell, which was found to be approximately 8.6μm, is pulling up the average. The second largest cell was found to be 4.9μm, and the remaining two cells were just over 3μm. The average width of the cells was 0.5μm. Campylobacter cells normally vary in length from 1.5-5μm and are around 0.5μm in width, this indicates that the 8.6μm cell is unusually long, but not unusually wide. ImageJ seems to be a very useful tool for analyzing cell dimensions.


ImageJ Project


Strategy: The goal was to count the number of people (players, coaches, umpires) on the field during a baseball game at Fenway Park.  When the image was analyzed, the range for number of pixels per person on the field had to be fairly small so that the people in the crowd and other objects were not included in the count.  Also, the shape of ellipses were used.

Data: It was found that 27 people were on the field during the game at Fenway.  The scale was set using the distance from home plate to first base as well (90 ft.).

Inferences: The count of people on the field was overestimated.  I believe this was due to the sizes of the players on the field compared to those in the stands.  It was difficult to determine an exact range of pixels for the program to classify as just players and not spectators at the game.