ClearedLeavesDB: an online database of cleared plant leaf images
© Das et al.; licensee BioMed Central Ltd. 2014
Received: 10 December 2013
Accepted: 19 March 2014
Published: 28 March 2014
Leaf vein networks are critical to both the structure and function of leaves. A growing body of recent work has linked leaf vein network structure to the physiology, ecology and evolution of land plants. In the process, multiple institutions and individual researchers have assembled collections of cleared leaf specimens in which vascular bundles (veins) are rendered visible. In an effort to facilitate analysis and digitally preserve these specimens, high-resolution images are usually created, either of entire leaves or of magnified leaf subsections. In a few cases, collections of digital images of cleared leaves are available for use online. However, these collections do not share a common platform nor is there a means to digitally archive cleared leaf images held by individual researchers (in addition to those held by institutions). Hence, there is a growing need for a digital archive that enables online viewing, sharing and disseminating of cleared leaf image collections held by both institutions and individual researchers.
The Cleared Leaf Image Database (ClearedLeavesDB), is an online web-based resource for a community of researchers to contribute, access and share cleared leaf images. ClearedLeavesDB leverages resources of large-scale, curated collections while enabling the aggregation of small-scale collections within the same online platform. ClearedLeavesDB is built on Drupal, an open source content management platform. It allows plant biologists to store leaf images online with corresponding meta-data, share image collections with a user community and discuss images and collections via a common forum. We provide tools to upload processed images and results to the database via a web services client application that can be downloaded from the database.
We developed ClearedLeavesDB, a database focusing on cleared leaf images that combines interactions between users and data via an intuitive web interface. The web interface allows storage of large collections and integrates with leaf image analysis applications via an open application programming interface (API). The open API allows uploading of processed images and other trait data to the database, further enabling distribution and documentation of analyzed data within the community. The initial database is seeded with nearly 19,000 cleared leaf images representing over 40 GB of image data. Extensible storage and growth of the database is ensured by using the data storage resources of the iPlant Discovery Environment. ClearedLeavesDB can be accessed at http://clearedleavesdb.org.
KeywordsDatabase Images Leaves Digital archiving Digital curation
The leaf organ is the site of the majority of primary production within land plants. Leaves vary in shape, composition, color, as well as the structure of their venation networks . Here, we focus on efforts to describe and analyze the last of these characteristics: the venation network. A leaf venation network represents the interconnected set of veins present in a leaf. Carbon, nutrients, and water are transported through xylem and phloem contained within the vascular bundles that constitute veins . In addition, veins confer material strength to the leaf due to the higher stiffness of veins relative to other leaf tissue [3, 4].
The analysis of leaf venation networks is of increasing interest due to multiple hypotheses linking venation networks to changes in leaf physiological rates [5, 6] and to the diversification of early land plants [7–9]. For example, increases in the density of leaf veins are hypothesized to increase physiological rates, such as net photosynthesis, because of the decreasing distance that water molecules must travel between the vein network, terminal sites of photosynthesis and stomates [10–12]. Similarly, the early adaptive radiation of land plants coincided with the appearance of increasingly reticulate (i.e. loopy) venation networks, as inferred from fossil leaves [13, 14]. Reticulate networks are known to confer added redundancy to leaves in the face of damage from herbivory or disease, providing a mechanism linking leaf venation structure to whole plant fitness .
In practice, the quantitative and qualitative analysis of venation networks requires the identification of veins as distinct from that of surrounding leaf tissue in areoles. Hence, a first step toward identifying a venation network is to “clear” a leaf [16, 17]. Multi-decadal efforts to collect cleared leaves have been undertaken by individual researchers and institutions. For example, in the USA there are now multiple, independent cleared leaf collections with thousands of specimens, e.g., held at the Smithsonian Institution , Yale Peabody Museum , New York Botanical Garden , and UC-Berkeley . In each case, digital images have been taken of the specimens and in many instances, institutional websites offer free access to all or a portion of these curated collections. However, these sites do not necessarily provide functionality for individual researchers unaffiliated with these institutions to store, manage, search, and analyze images online – indeed, such functionality is outside of the scope of each individual collection. There are also other generalized repositories of images of plants and plant organs such as CalPhoto , Morphbank , USDA Plant database , Encyclopedia of Life records  and Bisque . Bisque is notable for its support for secure image storage, analysis and data management capabilities. Bisque is also hosted on the iPlant  environment giving users control over sharing of images and analysis results. While these repositories are freely accessible, they are quite general in scope and are not designed for the specific set of features and meta-data associated with cleared leaf images. Moreover, individual collection sites are not linked, they do not share a common platform to share data, and they do not enable the integration of cleared leaf images with processed images and analysis. Hence there is a need for a complementary approach to enhance the utility of pre-existing digital-physical collections: a trans-institutional online database by which individual research groups and institutions can store, view, and share their cleared leaf images with the community.
In order to address these issues, we have developed the Cleared Leaf Image Database (ClearedLeavesDB). ClearedLeavesDB enables researchers to contribute, access, and share cleared leaf images. The database houses existing curated collections, such as a major contribution from the Smithsonian Institution collection, and enables researchers to contribute small-scale curated collections for dissemination, viewing and re-use. The web interface of ClearedLeavesDB is built on open-source technologies, allowing users to: (i) create and manage leaf image collections; (ii) upload/download images and their metadata in single or batch mode; (iii) annotate images; (iv) search images; and (v) link analysis data with the original images. The database provides varying degree of access based on user privileges which enhances data quality and reliability. As we describe below, we have developed ClearedLeavesDB to accelerate basic science and public understanding of plant biology generally, and leaf veins specifically.
Construction and content
We describe a user-oriented view of the ClearedLeavesDB system, given that our focus in this paper is on the practical usage of the web application by and for plant biologists. Major sections of the application are described below, including how to create and manage image collections, search and browse images, manage and download marked lists of images, and user roles. For users interested in viewing and modifying the source code, we note that the application is developed using Drupal ; a content management system built upon Apache web server , MySQL  database server and PHP .
Create and manage collections
Search and browse images
Interface for linking processed data with original images
ClearedLeavesDB is capable of storing original and processed images along with their metadata. The types of processed images might be binary images resulting from manual or automated segmentation of leaf veins from the leaf background as in [5, 6], skeletonized images of the approximate center-line of the leaf venation network , or aereole segmentation of the image . Additional processed data might include the results of quantitative analysis of the structure of the cleared leaf. In order to link post-processed images and data to the original image, we provide tools to upload processed images and its analysis data to the database from a user’s computer through a command line interface and application programming interface (API).
CSV file name (String data type)
Database user name (String data type)
Database user password (String data type)
The following Java code snippet can be used to invoke the API:
UploadProcessedImages.ToCLID(csvFile, username, password);
Please note that the CSV file should be compatible to the format as mentioned on the website. A sample file can be downloaded from ClearedLeavesDB website (http://www.clearedleavesdb.org/sites/default/files/sample_processed_upload.csv).
Analyzing images with third party software such as LeafGUI , LAMINA  and LIMANI  is enabled by the integrated batch download functionality. After analysis, processed images and their corresponding analysis data can be uploaded. These uploaded data are associated with the original image, hence heterogeneous and non-standard processing types can be uploaded, though the processed analysis data cannot yet be searched, a trade-off given the design decision to allow greater flexibility in allowing users to associate processed data with the original images.
List of user roles and their associated permissions
Browse collections and images
View collections and its metadata
View images and its metadata
View high resolution original images
Create image collections
Upload images to the collection
Bulk upload images to the collection
Upload metadata to the collection
Comment on collections and images
Maintain a marked list of images
Download images and its metadata from the marked list
Image upload quota
Add processed images and results
In this section we describe several examples to highlight the main use cases of our database: database browsing by a guest user; creating an image collection; managing collection membership, roles and permissions; and uploading processed images through the REST Web Services client.
A guest browsing the database
A user accessing the database will first arrive on the home page displaying the welcome message and a list of public collections available on the database. The user can either select any one of the collections shown or click on the “Collections” menu to search for a public collection and then click on it to browse the collection. The collection view (Figure 3) displays thumbnail images in the collection in a grid format along with members of the collection and comments posted by different users. The guest user does not have rights to post comments, add images or request membership to the group. Guests can select an image and view its metadata but cannot view full size images, post comments on an image or edit metadata. Guests can click on the “Images” menu to search for images with certain filters. From the image search page user can also navigate to the image details view page as described earlier. Guests do not have access to the marked list functionality i.e. adding images to the marked list from collection or search page.
Registered user and curator creating image collection
Here we describe the difference in creating an image collection by an authenticated user and a curator. Both these user roles have the rights to create image collections but registered users do not have rights to bulk upload images to the collection. We demonstrate this functionality using a registered user and a curator. Both these users have access to the “Create Collection” sub-menu in “Collections” menu on the database but the registered user does not have access to “Bulk Upload Images” sub-menu. After creating a collection, the registered user is redirected to the single image upload page whereas the curator is redirected to the bulk upload page.
Managing collection membership, roles and permissions
The founding curator (i.e., “author”) of a collection has all the rights to manage its members, roles and permissions. The author can visit the “Group” tab and click on the “Add People” link to add people to the collection and click on the “People” link to manage membership. Similarly “Roles” and “Permissions” link can be followed to manage roles and permissions of the collection.
Uploading processed images through REST web services client
ClearedLeavesDB is seeded with five distinct collections. The largest is from the National Cleared Leaf Collection housed at the Museum of Natural History, Smithsonian Institution, including specimens cleared and stained under the direction of Jack A. Wolfe, U.S. Geological Survey. The other collections are of (i) Arabidopsis thaliana specimens from a study of variation in venation network traits for which data include ecotypes, RILs, NILs, and vascular patterning mutants (courtesy of Benjamin Blonder, unpublished); (ii) specimens from multiple Populus tremuloides clones from the Colorado Rocky Mountains ; (iii) specimens taken from the University of Arizona arboretum ; (iv) specimens taken from oak trees of different species on the campus of the Georgia Institute of Technology (courtesy of Charles Price, unpublished).
Discussions and conclusions
ClearedLeavesDB provides a range of practical tools to store, manage and access cleared leaf images. The web interface for the database is built on open source technologies and is freely accessible online. At present, the database is seeded with over 40 GB of primary images of cleared leaves representing a total of 19,000 images. In doing so, ClearedLeavesDB provides a means to connect researchers, institutional repositories, and the public in accessing, sharing, and analyzing the biology of plant leaves. In this sense, ClearedLeavesDB is complementary to pre-existing websites that enable access to images of plants and plant organs, and specialized repositories of cleared leaf images. ClearedLeavesDB aims to bridge the gap between these two types of platforms by offering individuals and institutions a common platform, built on open source technology, to store, manage, share, view and analyze cleared leaf images. At present, ClearedLeavesDB leverages the infrastructure of iPlant  for flexible storage and access of third-party developed tools. In moving forward, we plan to extend the current system to enable further, integrated analysis of cleared leaf images and associated trait data. First, we plan to enable the association of images stored on ClearedLeavesDB with trait data pertaining to the original leaf and plant specimens, e.g., as stored on TRY-db.org , a global database for plant traits. Second, we plan to extend the current system to integrate the database with iPlant’s Data Store  built using an integrated rule-oriented data-management system (iRODS) . Hence, future efforts to analyze large-scale datasets of cleared leaves may benefit from bringing the software analysis tools to the data, rather than the other way around, as well as leveraging many of the other benefits of a scalable infrastructure.
Availability and requirements
The database is named as “ClearedLeavesDB” and is accessible at http://clearedleavesdb.org. The database can be accessed through any web browser; however it has been tested on Firefox (Version 27.0.1), Chrome (Version 33.0.1750.117), Safari (Version 6.1.1 and Version 7.0.1) and Internet Explorer (Version 11.0.9.x). A quick start guide is available on the website and is included here as Supplementary File 1. The public datasets present in the database are freely accessible but carry copyright information defined by the collection author.
Availability of supporting data
The data sets supporting the results of this article are available in the ClearedLeavesDB repository, http://clearedleavesdb.org.
Representational State Transfer
Java Virtual Machine.
The authors thank Scott Wing and Benjamin Blonder for their contributions of datasets to the initial release of Cleared Leaves DB and also thank Troy Hilley for technical contributions to database testing. This project was supported, in part, by a seed grant from the Center for Data Analytics at the Georgia Institute of Technology (to AB and JSW). Joshua S Weitz, PhD, holds a Career Award at the Scientific Interface from the Burroughs Wellcome Fund. CAP is supported by a Discovery Early Career Research Award from the Australian Research Council.
- Ellis B, DCD , Hickey LJ, Johnson KR, Mitchel JD, Wilf P, Wing SL: Manual of Leaf Architecture. 2009, Ithaca, NY: Cornell University Press, 978-0801475184Google Scholar
- Sack L, Holbrook NM: Leaf hydraulics. Annu Rev Plant Biol. 2006, 57: 361-381. 10.1146/annurev.arplant.56.032604.144141.View ArticlePubMedGoogle Scholar
- Niinemets U, Portsmuth A, Tobias M: Leaf size modifies support biomass distribution among stems, petioles and mid-ribs in temperate plants. New Phytol. 2006, 171: 91-104. 10.1111/j.1469-8137.2006.01741.x.View ArticlePubMedGoogle Scholar
- Niinemets U, Portsmuth A, Tobias M: Leaf shape and venation pattern alter the support investments within leaf lamina in temperate species: a neglected source of leaf physiological differentiation?. Funct Ecol. 2007, 21: 28-40.View ArticleGoogle Scholar
- Brodribb TJ, Feild TS: Stem hydraulic supply is linked to leaf photosynthetic capacity: evidence from New Caledonian and Tasmanian rainforests. Plant Cell and Environment. 2000, 23: 1381-1388. 10.1046/j.1365-3040.2000.00647.x.View ArticleGoogle Scholar
- Brodribb TJ, Feild TS, Jordan GJ: Leaf maximum photosynthetic rate and venation are linked by hydraulics. Plant Physiol. 2007, 144: 1890-1898. 10.1104/pp.107.101352.PubMed CentralView ArticlePubMedGoogle Scholar
- Boyce CK, Brodribb TJ, Feild TS, Zwieniecki MA: Angiosperm leaf vein evolution was physiologically and environmentally transformative. Proceedings of the Royal Society B-Biological Sciences. 2009, 276: 1771-1776. 10.1098/rspb.2008.1919.PubMed CentralView ArticleGoogle Scholar
- Brodribb TJ, Feild TS: Leaf hydraulic evolution led a surge in leaf photosynthetic capacity during early angiosperm diversification. Ecol Lett. 2010, 13: 175-183. 10.1111/j.1461-0248.2009.01410.x.View ArticlePubMedGoogle Scholar
- Feild TS, Brodribb TJ, Iglesias A, Chatelet DS, Baresch A, Upchurch GR, Gomez B, Mohr BA, Coiffard C, Kvacek J, Jaramillo C: Fossil evidence for Cretaceous escalation in angiosperm leaf vein evolution. Proc Natl Acad Sci U S A. 2011, 108: 8363-8366. 10.1073/pnas.1014456108.PubMed CentralView ArticlePubMedGoogle Scholar
- Price CA, Wing S, Weitz JS: Scaling and structure of dicotyledonous leaf venation networks. Ecol Lett. 2012, 15: 87-95. 10.1111/j.1461-0248.2011.01712.x.View ArticlePubMedGoogle Scholar
- Raven JA: The evolution of vascular plants in relation to quantitative functioning of dead water-conducting cells and stomata. Biol Rev Camb Phil Soc. 1993, 68: 337-363. 10.1111/j.1469-185X.1993.tb00735.x.View ArticleGoogle Scholar
- Sack L, Tyree MT, Holbrook NM: Leaf hydraulic architecture correlates with regeneration irradiance in tropical rainforest trees. New Phytol. 2005, 167: 403-413. 10.1111/j.1469-8137.2005.01432.x.View ArticlePubMedGoogle Scholar
- Kull U, Herbig A: Leaf venation of angiosperms - form and evolution. Naturwissenschaften. 1995, 82: 441-451. 10.1007/BF01131595.View ArticleGoogle Scholar
- Sack L, Scoffoni C, McKown AD, Frole K, Rawls M: Developmentally based scaling of leaf venation architecture explains global ecological patterns. Nat Commun. 2012, 3: doi:10.1038/ncomms1835Google Scholar
- Sack L, Dietrich EM, Streeter CM, Sanchez-Gomez D, Holbrook NM: Leaf palmate venation and vascular redundancy confer tolerance of hydraulic disruption. Proc Natl Acad Sci U S A. 2008, 105: 1567-1572. 10.1073/pnas.0709333105.PubMed CentralView ArticlePubMedGoogle Scholar
- Bates JC: A method for clearing leaves. Am Nat. 1931, 65: 288-288. 10.1086/280373.View ArticleGoogle Scholar
- Gardner RO: Overview of botanical clearing technique. Stain Tech. 1975, 50: 99-105.Google Scholar
- Smithsonian Institution. 2014,http://www.si.edu/,
- Yale Peabody Museum. 2014,http://peabody.yale.edu/,
- New York Botanical Garden. 2014,http://www.nybg.org,
- UC Berkeley Cleared Leaf Collections. 2014,http://www.ucmp.berkeley.edu/collections/plants/clearedleaf.html,
- CalPhotos: A database of photos of plants, animals, habitats and other natural history subjects. 2014, Berkeley: BSCIT, University of California,http://calphotos.berkeley.edu/,Google Scholar
- Morphbank: Biological Imaging. 2014, USA: Florida State University, Department of Scientific Computing, Tallahassee, FL, 32306-4026.http://www.morphbank.net,Google Scholar
- The PLANTS Database: USDA, NRCS. 2014, NC 27401-4901USA: National Plant Data Team, Greensboro, 27401-4901.http://plants.usda.gov,Google Scholar
- Encyclopedia of Life. 2014,http://www.eol.org,
- Kvilekval K, Fedorov D, Obara B, Singh A, Manjunath BS: Bisque: a platform for bioimage analysis and management. Bioinformatics. 2010, 26: 544-552. 10.1093/bioinformatics/btp699.View ArticlePubMedGoogle Scholar
- Jorgensen RA, Stein L, Rain S, Andrews G, Chandler V: The iPlant collaborative: a cyberinfrastructure-centered community for a new plant biology. In Vitro Cellular & Developmental Biology-Animal. 2008, 44: S26-S26.Google Scholar
- The Apache HTTP Server.http://httpd.apache.org/,
- MySQL Database Server.http://www.mysql.com/,
- Price CA, Symonova O, Mileyko Y, Hilley T, Weitz JS: Leaf extraction and analysis framework graphical user interface: segmenting and analyzing the structure of leaf veins and areoles. Plant Physiol. 2011, 155: 236-245. 10.1104/pp.110.162834.PubMed CentralView ArticlePubMedGoogle Scholar
- Bylesjo M, Segura V, Soolanayakanahally RY, Rae AM, Trygg J, Gustafsson P, Jansson S, Street NR: LAMINA: a tool for rapid quantification of leaf size and shape parameters. Bmc Plant Biology. 2008, 8: 82-10.1186/1471-2229-8-82. doi:10.1186/1471-2229-8-82PubMed CentralView ArticlePubMedGoogle Scholar
- Dhondt S, Van Haerenborgh D, Van Cauwenbergh C, Merks RMH, Philips W, Beemster GT, Inzé D: Quantitative analysis of venation patterns of Arabidopsis leaves by supervised image analysis. Plant J. 2012, 69: 553-563. 10.1111/j.1365-313X.2011.04803.x.View ArticlePubMedGoogle Scholar
- Blonder B, Violle C, Enquist BJ: Assessing the causes and scales of the leaf economics spectrum using venation networks in Populus tremuloides. J Ecol. 2013, 101: 981-989. 10.1111/1365-2745.12102.View ArticleGoogle Scholar
- Blonder B, Violle C, Bentley LP, Enquist BJ: Venation networks and the origin of the leaf economics spectrum. Ecol Lett. 2011, 14: 91-100. 10.1111/j.1461-0248.2010.01554.x.View ArticlePubMedGoogle Scholar
- Kattge J, Díaz S, Lavorel S, Prentice IC, Leadley P, Bönish G: TRY – a global database of plant traits. Glob Chang Biol. 2011, 17: 2905-2935. 10.1111/j.1365-2486.2011.02451.x.PubMed CentralView ArticleGoogle Scholar
- iRODS (Integrated Rule-Oriented Data System.http://www.irods.org,
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.