Autarchy of the Private CaveScience »

Academia or life?

Bogdan — Sat, 16 Apr 2011 10:56:42 +0000

Worth reading: Goodbye academia, I get a life.

Amazonia! 6462 human microarray datasets

Bogdan — Sun, 06 Mar 2011 19:18:51 +0000

Amazonia! – explore the jungle of microarray results

Paradoxically, the tremendous downpour of microarray results prevents a simple use of expression data. Therefore, we propose a thematic entry to public transcriptomes: you may for instance query a gene on a “Stem Cells page”, where you will see the expression of your favorite gene across selected microarray experiments related to stem cell biology. This selection of samples can be customized at will among the 6462 samples currently present in the database.

Every transcriptome study results in the identification of lists of genes relevant to a given biological condition. In order to include this valuable information in any new query in the Amazonia! database, we indicate for each gene in which lists it is included. This is a straightforward and efficient way to synthesize hundreds of microarray publications.
A special feature of Amazonia! is the field of human stem cells, notably embryonic stem cells.

Introduction to Python for bioinformatics

Bogdan — Fri, 25 Feb 2011 12:03:55 +0000

This overview presentation is two years old, but still a highly valuable resource: modules and tools mentioned are alive and useful.
I think this is the second presentation by Giovanni I’m embedding (first one being about GNU/make for bioinformatics).

Introduction to python for bioinformatics

How to replace newlines with commas, tabs etc (merge lines)

Bogdan — Tue, 16 Nov 2010 08:20:45 +0000

Imagine you need to get a few lines from a group of files with missing identifier mappings. I have a bunch of files with content similar to this one:

ENSRNOG00000018677 1368832_at 25233
ENSRNOG00000002079 1369102_at 25272
ENSRNOG00000043451 25353
ENSRNOG00000001527 1388013_at 25408
ENSRNOG00000007390 1389538_at 25493

In the example above I need '25353', which does not have corresponding affy_probeset_id in the 2nd column.

It is clear how to do that:

PLAIN TEXT

CODE:

sort -u *_affy_ensembl.txt | grep -v '_at' | awk '{print $2}'

This outputs a column of required IDs (EntrezGene in this example):

116720
679845
309295
364867
298220
298221
25353

However, I need these IDs as a comma-separated list, not as newline-separated list.

There are several ways to achieve the desired result (only the last pipe commands differ):

PLAIN TEXT

CODE:

sort -u *_affy_ensembl.txt | grep -v '_at' | awk '{print $2}' | gawk '$1=$1' ORS=', '

PLAIN TEXT

CODE:

sort -u *_affy_ensembl.txt | grep -v '_at' | awk '{print $2}' | tr '\n' ','

PLAIN TEXT

CODE:

sort -u *_affy_ensembl.txt | grep -v '_at' | awk '{print $2}' | sed ':a;N;$!ba;s/\n/, /g'

PLAIN TEXT

CODE:

sort -u *_affy_ensembl.txt | grep -v '_at' | awk '{print $2}' | sed ':q;N;s/\n/, /g;t q'

PLAIN TEXT

CODE:

sort -u *_affy_ensembl.txt | grep -v '_at' | awk '{print $2}' | paste -s -d ","

These solutions differ in efficiency and (slightly) in output. sed will read all the input into its buffer to replace newlines with other separators, so it might not be best for large files. tr might be the most efficient, but I haven't tested that. paste will re-use delimiters, so you cannot really get comma-space ", " separation with it.

Sources: linuxquestions 1 (explains used sed commands), linuxquestions 2, nixcraft.

Overlaying gene expression data onto pathways from databases

Bogdan — Fri, 05 Nov 2010 13:20:06 +0000

Superimposing gene expression data onto pathways from databases is a common task in the final steps of microarray data analysis - that is, biological interpretation and results discussion.

I have found many tools which claim to facilitate this procedure. Some of them are reviewed below (in no specific order).

Pathway Explorer by Bernhard Mlecnik was last updated in 2007, but is fully functional (I believe it is being maintained without changes to the last-updated date). Both online and downloadable Java applications are available. Note that for downloadable application you will need to obtain a license key - the procedure is well documented and was very fast for me.

Pathway Explorer supports import from 3 sources: KEGG xml files, biocarta URLs, and GenMAPP URLs. Import from KEGG does work as described in the short manual, and seems functional (I had some problems exporting/saving the resulting picture, but didn't investigate further). Biocarta import seems to work, but for some reason does not display expression levels of pathway components. I could not test the import of GenMAPP pathways, because they are not available online.

I found Pathway Explorer good, but then switched to PathVisio (reviewed next), because for some reason Зathway Explorer was recognising only a small fraction of genes from my expression data. It could be that identifiers mappings are outdated, but this is just a guess.

PathVisio appears to be a spin-off of GenMAPP and WikiPathways. It excells at importing/visualizing WikiPathways data, which even comes bundled with PathVisio Java application. It is easier to use than Pathway Explorer, and it seems to recognize more genes (although still not all the genes which are present in the data). There is KEGG pathways support, but it is not always usable - many edges (links between genes/proteins) are absent, so instead of a pathway you get a bunch of nodes relevant to a pathway, but cannot really see how they are connected. PathVisio supports an insanely long list of database identifiers, so it is highly unlikely that you will have to map your data to use a different identifier. This pathway mapper exports to several formats, including PNG and PDF.

I could not fully test AffyWEB, because it doesn't list rat arrays we used. Trying their barley genome example did work, so the tool is probably functional. It overlays your expression data onto KEGG pathways.

G-language Microarray System is a comparatively simple pathway visualizer. It accepts CSV files containing EntrezGene IDs column with a single column of expression values normalized to 1-100 range, fetches requested KEGG pathway, and generates a Flash (SWF) object depicting that pathway with coloured components. It does work with sample data. I was too lazy to normalize my expression data to [1;100] range, and SWF is not exactly a usable format, so I haven't tested this tool any further (you can right-click to zoom in the flash pathway below).

If time permits (or work requires) this post may be extended with the reviews of GenMAPP, GEPAT, KEGG2heatmap script, EGAN, MapMan, Pathway Miner, ArrayXPath, VisANT, SpotXplore, and maybe others.

Please comment to share your experience using pathway expression overlaying tools or to suggest other tools.

Batch-retrieve EntrezGene homologs using NCBI’s HomoloGene and R’s annotationTools

Bogdan — Wed, 27 Oct 2010 10:49:01 +0000

Install the annotationTools R package:
source("http://bioconductor.org/biocLite.R")
biocLite("annotationTools")
Download full HomoloGene data file from ftp://ftp.ncbi.nlm.nih.gov/pub/HomoloGene/current
library(annotationTools)
homologene = read.delim("homologene.data", header=FALSE)
mygenes = read.table("file with one entrez ID of the source organism per line.txt")
getHOMOLOG(unlist(mygenes), taxonomy_ID_of_target_organism, homologene) [alternatively, wrap the call to getHOMOLOG into unlist to get a vector]

It might be easier to achieve the same results with a Perl script calling NCBI's e-utils.

International salary survey in sciences (2010)

Bogdan — Thu, 14 Oct 2010 15:28:42 +0000

Nature published the said survey based on responses of over 10000 employees in science. It has lots of multi-axis data to explore, and some major trends are discussed in the special report. Highly recommended for anyone considering science career changes.

Tools for conversion of IDs in genomics

Bogdan — Tue, 10 Aug 2010 12:31:44 +0000

Tools for conversion of IDs in genomics

Вірний переклад українською терміну microarray (post in Ukrainian)

Bogdan — Wed, 26 May 2010 12:12:07 +0000

Викладену нижче власну позицію вважаю найправильнішою (і використовую з 2007 року).

Префікс мікро в українській мові є, і позначає певну кратність (10^-6) числової величини (а також просто щось маленьке) – тому його можна зберегти при перекладі першої половини складного слова microarray. Цей префікс також входить до системи одиниць СІ.

А от слова арей (як вживають деякі автори) в українській мові немає. Також немає сенсу його запозичувати, оскільки існують переклади (слова-еквіваленти). Один зі словників пропонує такі варіанти перекладу слова array українською (у різних контекстах):

множина, набір, комплект
розташування, решітка, сітка
масив, список, поле, ряд
решітка даних
масив даних
матриця

Я пропоную використовувати термін мікромасив (та похідний від нього мікромасив-експеримент). Цей термін має перевагу над вживаним у Російській Федерації "микрочип-экспериментом", оскільки "мікрочіп" або просто "чіп" – це усталений термін електроніки, де він позначає кремнієвий електронний елемент з високим ступенем упаковки напівпровідників; натомість "масив" - це і набір/список [даних], и [двомірна] матриця [даних/ознак/зондів/будь-чого]. Відповідно, мікромасив – це маленька матриця або маленький набір [олігонуклеотидних/кДНК зондів]. Додатковим аргументом проти використання слова чіп вважаю його запозиченість.

Використання інших варіантів перекладу слова array або не відповідає суті об'єкту, або має неоднозначне трактування. Наприклад, мікроматриця: в молекулярній біології матриця - це ланцюг ДНК, з якого іде синтез, а в ширшому значенні - взагалі будь-яка модель, з якої виготовляють зразки. (Звичайно, слово матриця також є синонімом слова масив у значенні двомірний масив / двомірна матриця, але слово масив не має - наскільки мені відомо - альтернативних трактувань у молекулярній біології). Розглядати інші варіанти перекладу слова array не вважаю за потрібне, оскільки вони ще менш вдалі за матрицю.

Таким чином, вірним перекладом терміну microarray є слово мікромасив.

R tutorial links

Bogdan — Mon, 29 Mar 2010 16:17:31 +0000

R time series tutorial (2010, a website of the "Time Series Analysis and Its Applications: With R Examples" book)
Statistics with R (2007)
R for programmers PDF (2008, 104 pages, linked to from here)
Brief R tutorial (2004)
Statistical computing with R: a tutorial (2004)
An introduction to R (from the official r-project website, should be always up-to-date)
R tutorial (date unknown, definitely newer than 2005)

R script to filter probesets with log-expression values below the lowest spike-in

Bogdan — Wed, 27 Jan 2010 12:44:02 +0000

Sometimes there is a need to remove all the probesets, which have expression values below the minimal spike-in intensity on the Affymetrix microarray. The reasoning behind this procedure is simple: minimal-expression spike-ins represent the bottom margin of microarray sensitivity, and anything below that margin cannot be reliably quantified - which also means that both fold-change and p-value of expression variance will be unreliable for these probesets.

Here's a simple R script to do just that. It is abundantly commented, and also contains an optional (commented out) fragment which allows the removal of more low-variance, low-intensity probesets.

Hint: click the "plain text" box header to be able to right-click the code, "Select All", and then "Copy".

PLAIN TEXT

CODE:

filter_below_spikes = function(eset) {
# Finds max(lowest) AFFX/spike-in intensity, and removes rows consisting entirely of values below max(lowest).
# @param eset
# ExpressionSet
# @returns
# exprs(ExpressionSet), filtered
# Without Biobase exprs() will not work.
require(Biobase)
expr = exprs(eset)
# 'expr' sample:
# 1
# 1367452_at 10.880208
# 1367453_at 10.554647
cat(nrow(expr), "rows before filtering.\n")
# Make a vector of spike row names.
spikes = grep("AFFX", rownames(expr), value = TRUE)
cat("Expression matrix has", length(spikes), "spike-in rows.\n")
cat("Summary of spike-in values distribution follows:\n")
print(summary(expr[spikes, ]))
# Find max(lowest) spike-in values.
minval_max = max(as.double(substr(grep("Min", summary(expr[spikes, ]), value = TRUE), 10, 14)))
cat("max(minimal spike-in log-intensity values) =", minval_max, "\n")
# Remove spike-ins from expr.
expr = expr[grep("AFFX", rownames(expr), value = TRUE, invert = TRUE), ]
nospike_rows = nrow(expr)
cat(nospike_rows, "rows remaining after the removal of", length(spikes), "spike-in probesets.\n")
# Optional: calculate max(SD) of all removed rows.
#bad_sds_max = max(apply(expr[!apply((expr> minval_max), 1, any),], 1, sd))
# Now remove all rows, where each value is <= minval_max.
expr = expr[!apply((expr <= minval_max), 1, all), ]
cat(nrow(expr), "rows remaining after filtering out", nospike_rows - nrow(expr), "probesets with all values below", minval_max, "\n")
#cat(bad_sds_max, "is max(SD) of all", nospike_rows - nrow(expr), "filtered probesets with all values below", minval_max, "\n")
# Optional: Remove *some* of the rows, which have at least one value below minval_max, and row_SD <= bad_sds_max.
#pre_final_rows = nrow(expr)
#expr = expr[(apply(expr, 1, sd)> bad_sds_max) | (apply(expr, 1, min)> minval_max), ]
#cat(pre_final_rows-nrow(expr), "rows with SD <=", bad_sds_max, "and min(row) <=", minval_max, "were removed.\n")
#cat(nrow(expr), "final rows returned.\n")
return(expr)
}

Sample use:

> source("script.R")
> expr.filtered = filter_below_spikes(eset)

Comments and suggestions are welcome.

DIYbio, biohackers, and Open Source Medicine

Bogdan — Sat, 25 Jul 2009 14:19:38 +0000

DIYbio is

an organization that aims to help make biology a worthwhile pursuit for citizen scientists, amateur biologists, and DIY biological engineers who value openness and safety.

DIYbio also has a google group, where a wide range of questions - from bio-patents to DIY gel electrophoresis shopping list and model organisms is dicussed. There is also a DIYbio/biohacking FAQ.

Today for me is the day of discoveries. I learned about the International Open Space Initiative (to give robotics enthusiasts a way to send their tele-controlled and/or intelligent robots to the Moon and Mars), about the DIYbio and biohackers, about OpenManufacturing (which doesn't seem to have produced enough content to link to), Open Source Medicine (ouch!), BioBrick Assembly Kit (with an assembly manual), OpenWetWare, and a whole bunch of other awesome and inspiring community efforts, which do not belong here.

Do you feel the wind of change?

SciVee.TV: YouTube for science

Bogdan — Wed, 15 Jul 2009 14:00:17 +0000

Stumbled upon SciVee.TV - an open video upload service for research-related videos.

I believe it is highly useful. Compare: watching an 8-10 minute video of someone's research to reading their article on that same subject. For me, those 8-10 minutes make video option a clear winner.

One of the envisioned uses of SciVee is to upload videos describing peer-reviewed published articles. This has two benefits for the reader: quickly getting acquainted with the essence of the article, and having that article as a complete reference for any questions not discussed in the video. For the author, this gives an additional bonus of higher visibility of his research.

Personally, I've immediately found 3 videos pertinent to my topic. Of those, one was accompayning an article in PloS Biology, one was an hour-long lecture, and one was a poor quality audio recording of someone's intended research.

SciVee is young, and that is currently the largest drawback: not much could be found in a narrow research field. But I'm sure it will grow.

10 simple rules for getting published

Bogdan — Fri, 26 Jun 2009 14:20:56 +0000

Recently, I have come across an excellent piece of advice called Ten simple rules for getting published. The only thing I have to add is that final rule #10 should be kept in mind while checking through all the previous rules - e.g. when editing someone's submission, make sure that you are in position to be the editor for that article, and make sure your decision will influence chief editor's decision - otherwise there's no use reviewing.

PLoS Computational Biology, where the "simple rules" were published, has a Ten simple rules collection, which includes a handful of other useful advice articles, like 10 simple rules for selecting a postdoctoral position.

GNU make for bioinformatics presentation

Bogdan — Fri, 05 Jun 2009 09:37:21 +0000

Giovanni Dall’olio has recently posted a presentation on using make.

Although it has "bioinformatics" on the title page, this is a good and very easy to understand make intro.

Original post is here.

Ensembl Genomes launches Protists, Bacteria and Metazoa

Bogdan — Tue, 21 Apr 2009 13:27:40 +0000

The following sites are available:

http://bacteria.ensembl.org
http://protists.ensembl.org
http://metazoa.ensembl.org

During summer, two more sites - for Fungi and Plants - should be made available.

Learn more about Ensembl Genomes project.

How to create custom Affymetrix CDF file

Bogdan — Mon, 23 Mar 2009 12:01:41 +0000

First, learn about custom CDFs and why they are needed.

The aroma.affymetrix R package google group has a how-to: create a CDF annotation file from scratch.

Also useful: how to convert CDF into an R package, which has all CDF data available (as a PDF with more details).

BioGPS: by the makers of Gene Expression Atlas

Bogdan — Sat, 22 Nov 2008 10:48:46 +0000

If you were a frequent user of GNF SymAtlas, then you'd better bookmark BioGPS - Your Gene Portal System. BioGPS is basically the same gene expression atlas, but with a completely different interface, and more flexible ideology (e.g. expression atlas is now just a "plugin", and more of those can be plugged in).

There are also some easter-egg-like features: try hovering the BioGPS logo in the top left corner several times...

Ensembl release 51 is out!

Bogdan — Thu, 20 Nov 2008 13:51:34 +0000

This in itself wouldn't be so exciting, if it were not for the new webcode! If you do visit Ensembl now, you will be definitely surprised with the page loading speed - it is amazingly faster than it used to be! Also, as you dig deeper and deeper, you'll see that there are a lot more other differences - starting with the new design, and not forgetting the changed page organization logic.

To cut the long story short, here's the list of changes in the new Ensembl 51 webcode release. Other changes to Ensembl in release 51 are also available.

One of the new features which caught my attention is the ability to add custom tracks in Ensembl (which is a long-available feature in UCSC Genome Browser). Interestingly, you do not even have to be logged in to use this feature. We shall be considering providing the TFBS custom track for several species, as predicted de-novo by our evolutionary conserved tfbs search tool (binding site finder), but this is a long shot, given the already published COTRASIF development roadmap.

There is one more great news which is kinda insufficiently highlighted: the brand-new Ensembl Genome Browser website, which (as planned, it hasn't yet started operations) will provide access not only to vertebrates, but also to other taxonomy groups. The full list is:

Metazoa
Protists
Bacteria
Plants
Fungi

Citation and reference managers for Linux: short overview

Bogdan — Wed, 29 Oct 2008 12:00:46 +0000

For some period of time, I had to switch to my Windows installation when writing articles with references, because I keep my references database there and there is an extremely convenient add-on for MS Office to insert citations and format bibliography on the fly.

However, it is quite inconvenient to make/run programs/scripts under Linux, and then switch to Windows to describe the obtained results in the article-like form. Thus, I briefly investigated available citation managers for Linux.

Note: this is just a review at the moment, I hadn't tried yet the options described.

First comes the Zotero+OpenOffice Writer, as suggested by sybille (edited quote):

Zotero includes support for an extension for OpenOffice.org Writer.
There's a lot of great documentation about Zotero at the home page for the extension, including various useful demos of its features.
If you want the newest version of Zotero, it's better to install it from that page rather than the official Firefox add-ons site.
Once you've installed Zotero, you can install the extension for OpenOffice.org as follows: first download the extension from the Zotero site. Unzip the file so that you have a file called Zotero.oxt somewhere where you can find it. Then open a new text document with ooWriter, go to Tools -> Extension Manager, choose "Add." Navigate to the directory where you put Zotero.oxt and choose it. Once the extension installs, restart Writer, and now you'll see a new toolbar for inserting references and bibliographies into your Writer documents.
I prefer Zotero because it makes it so easy to enter bibliographical information into the database it manages. Generally it's just a matter of locating the publication or resource you want to cite in an online library or journal service (some newspapers are currently supported as well), and then clicking on an icon in the Firefox address bar. Maybe some small edits are needed depending on how the information is imported, but that's it.
Zotero also allows taking notes and adding/annotating attachments - links to files, copies of files, snapshots of web pages, etcetera etcetera. There are some new full-text search capabilities as well.
The OpenOffice extension does the same kind of job as Bibus, using UNO service to insert citations and bibliographies from Zotero into Writer documents in any number of styles (currently over 1100). These references can be updated when the Zotero database changes.
I'm finding the Firefox/Zotero combination very useful for my research, and I think this setup is worth a look for anyone doing research.

Zotero summary: easy references database maintenance and import into; has bibliography formatting capabilities with over 1100 styles; has OpenOffice extension; seems to have RefMan/EndNote import feature (not verified). This is the first candidate to be tried, even without looking into other options.

Bibus:

Bibus is a bibliographic and reference management software. As with other such tools, Bibus allows one to search, edit, and sort bibliographic records. In addition, Bibus contains features that makes it unique among open source and even commercial bibliographic databases:
* Hierarchical organization of the references with user defined keys
* Designed for multiuser environments
o You can share the database between an "unlimited" number of users
o Each user will have its own classification
o You can define read-only and read-write users
* Live queries; that is searches that update as the database changes
* On-line PubMed queries
* On-line eTBLAST queries
* Insertion of references and formatting of bibliographies into two widely used Word Processors (OpenOffice.org and Microsoft Word)
* Foreign language support through Unicode and gettext. As of version 1.4, Bibus is available in English, Chinese, Czech, French, German, Hungarian, Portuguese, Slovenian, Spanish.
* Bibus works on most modern platforms:
o Gnu/linux with OpenOffice.org
o Windows98/2000/XP with OpenOffice.org and Microsoft Word
o On other platforms (MacOS X, ...) Bibus works with Microsoft Word but styles cannot be created at this point (this should change soon for MacOS X)

Bibus summary: single integrated solution for managing references and citations. Based on the description only, looks on par with Zotero (if not superior to). However, nothing is said about the ease of importing new citations. Has Debian package, so this is already installed and ready to try it out (simultaneously with Zotero, which is now also installed). One more theoretical benefit of Zotero might be in it's cross-computerability, meaning that simple synchronization of FireFox add-ons might also synchronize your references, which is a great plus (need to investigate that). Bibus, on the other hand, uses local SQLite database, which adds bits of hassle when there is a need to keep several workstations in sync.

Jorge:

Jorge is an OpenSource bibliography management software entirely written in Java. It will provide an free alternative to professional bibliography management software like EndNote©.
In the first step you will be able to create your own literature database. From this literature database you can create citation lists and export them to BibTeX or plain text formats. The database itself will be an XML-File as the saveable citation lists will be.
In a second step I try to implement an search interface to the Z39.50 protocol libraries. This is for to connect to public libraries via the ISO Z39.50 protocol. With this search you will be able to access a wide variety of libraries via the internet and import search results to your local database.

Jorge summary: has an interesting motivation behind the name of the software; I dislike Java-software, so won't try this one until others prove to be insufficient.

Referencer

Referencer is a Gnome application to organise documents or references, and ultimately generate a BibTeX bibliography file. Referencer includes a number of features to make this process easier:
* Smart web links
Referencer uses documents' metadata to provide handy links to the document's web location — no need to maintain your own bookmarks.
* Import from BibTeX, Reference Manager and EndNote
No need to start from scratch — Referencer will import your existing bibliography files using the BibUtils library.
* Tagging
No need to organise your documents into rigid directory trees — with Referencer you can use tags to categorise your documents.
* Automatic arXiv, PubMed and CrossRef metadata retrieval
If you show Referencer a PDF which has an arXiv ID or DOI code, Referencer will retrieve the metadata for this document over the internet.
* Python plugin support
Referencer can be extended using the versatile Python scripting language.
* Localisation
Referencer is translated into many languages.

Referencer summary: can insert citations into LyX documents; uses BibTex, which I believe is a good thing ; has tagging and import from online sources; is present in Debian testing repository; is suitable for those used to working with LaTex; this is an option to try, but for me personally requires time investments into learning Lyx/LaTex/BibTex, so this goes after Zotero and Bibus in my personal rating (it would go first, if I had LaTex experience).

JabRef:

JabRef is an open source bibliography reference manager. The native file format used by JabRef is BibTeX, the standard LaTeX bibliography format. JabRef runs on the Java VM (version 1.5 or newer), and should work equally well on Windows, Linux and Mac OS X.
BibTeX is an application and a bibliography file format written by Oren Patashnik and Leslie Lamport for the LaTeX document preparation system. General information about BibTeX.
Bibliographies generated by LaTeX and BibTeX from a BibTeX file can be formatted to suit any reference list specifications through the use of different BibTeX style files. We support this initiative to build a searchable database of BibTeX style files, organized by journal names: LaTeX bibliography style database.
# Detailed editing of BibTeX entries.
# Search a pattern in the whole bibliography.
# You can group entries explicitly, by keywords or any other fields.
# Import: BibTeXML, CSA, Refer/Endnote, ISI Web of Science, SilverPlatter, Medline/Pubmed (xml), Scifinder, OVID, INSPEC, Biblioscape, Sixpack, JStor and RIS.
# Export: HTML, Docbook, BibTeXML, MODS, RTF, Refer/Endnote and OpenOffice.org.
# You can add your own fields to any BibTeX entry type.
# Launch external applications: PDF/PS viewers, web browser, insert citations into LyX, Kile, LatexEDitor, Emacs, Vim and WinEdt
# Search Medline, Citeseer, IEEEXplore and arXiv
# Support for XMP Metadata in PDFs
# Improve the workflow of sharing PDFs and bibliography information
# Plugin functionality

JabRef summary: similar to Referencer, but in Java. Would fit cross-platform BibTex-ers. Will not be further considered.

Sixpack: last updated in 2004, listed for completeness.

Feel free to comment about your best experience with reference/citation managers on Linux. Also, feel free to suggest new entries for the list above.