Opportunities at the Intersection of Bioinformatics and Health Informatics A Case Study Perry L. Miller, MD, PhD

Opportunities at the Intersection of Bioinformatics and Health Informatics
A Case Study
Perry L. Miller, MD, PhD

This paper provides a “viewpoint discussion” based on a presentation made to the 2000 Symposium of the American College of Medical Informatics. It discusses potential opportunities for researchers in health informatics to become involved in the rapidly growing field of bioinformatics, using the activities of the Yale Center for Medical Informatics as a case study. One set of opportunities occurs where bioinformatics research itself intersects with the clinical world.
...
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC79037/

Evolving research trends in bioinformatics
Carolina Perez-Iratxeta, Miguel A. Andrade-Navarro and Jonathan D. Wren

Corresponding author. Jonathan D. Wren, PhD, The University of Oklahoma, 101 David L. Boren Blvd., Rm. 2025, Norman, OK 73019, USA. Tel: +1-405-325-3415; Fax: +1-405-325-3442; E-mail: Jonathan.Wren@OU.edu
http://bib.oxfordjournals.org/cgi/content/full/8/2/88

http://bib.oxfordjournals.org/current.dtl#PAPERS

http://bib.oxfordjournals.org/cgi/content/full/10/6/593

Grad School Advice

http://web.pdx.edu/~obrienk/gradschool%20advice.pdf

http://www.howigotintostanford.com/shorter.php

http://www.justcolleges.com/grad/why-graduate-school.htm

http://www.military.com/finding-a-school/why-graduate-school

http://genomemedicine.com/content

http://www.biomedcentral.com/bmcbioinformatics?page=2&relevant=true

One of the great challenges of plant research is protecting crops from pathogenic attack. Soderlund describes a range of large-scale experimental studies for the analysis of plant–pathogen interactions, describing the bioinformatics approaches and tools that have been influential in gaining new knowledge from these data sets. She distinguishes between those problems that may be solved using generic bioinformatics methodologies and those for which bespoke solutions are required. Soderlund discusses key resources including fungal and oomycete databases and initiatives such as the PAMGO ontology, emphasizing the role of improved web-based resources in the knowledge discovery process.

Next, we shift our focus towards the important theme of understanding the genetic basis of plant traits. Zhang et al. describe methodologies for inferring genotype–phenotype associations through whole genome studies, focusing on the use of mixed model. The authors show that although many parts of the analytical process are rooted in animal studies and while some transfer directly to plant studies, others do not and need to be adapted. They discuss challenges to software engineering of the various methodologies involved, particularly in light of new high-throughput marker data sets, and discuss a number of software tools with regard to many key user requirements.

Finally, Lysenko et al. describe a problem that has arisen due to our capability to develop the data sets such as those described throughout this issue, that of developing software systems to integrate across information resources. The authors describe some of the problems inherent to data integration and discuss and contrast methods which are being used to solve them, while also noting the benefits of such approaches in generating new knowledge. They discuss two distinct, plant-oriented data integration case studies in the model plant Arabidopsis thaliana with reference to their own Ondex system, and look to new challenges in the form of high-throughput data sets, standards and ontologies.
http://bib.oxfordjournals.org/cgi/content/full/10/6/593

Disease to gene

http://bib.oxfordjournals.org.proxy.lib.sfu.ca/cgi/content/full/10/6/654
The Fungal Genome Initiative (FGI) [3] was formed to sequence fungal genomes,

Therefore, various software packages have been developed to partition the ESTs based on their nucleotide composition. Huitema et al. [35] used the fact that the GC content can sometimes be quite different between the host and pathogen. Maor et al. [36] developed the PF-IND method that uses codon usage bias. The problem with using codon usage is the need to find the in-frame coding sequence, and the fact that ESTs are also composed of untranslated regions. Therefore, Emmersen et al. [37] used triplet frequencies which are computed over both the coding and non-coding regions. The triplet frequencies are calculated using a sliding window over a training set of plant and pathogen ESTs, and a SVM (support vector machine) is used for the classifications.

AI - ch. 4 part 2

Annealing, in metallurgy and materials science, is a heat treatment wherein a material is altered, causing changes in its properties such as strength and hardness. It is a process that produces conditions by heating to above the re-crystallization temperature and maintaining a suitable temperature, and then cooling. Annealing is used to induce ductility, soften material, relieve internal stresses, refine the structure by making it homogeneous, and improve cold working properties.

In the cases of copper, steel, silver, and brass this process is performed by substantially heating the material (generally until glowing) for a while and allowing it to cool slowly. In this fashion the metal is softened and prepared for further work such as shaping, stamping, or forming.

Simulated annealing - I guess that explains why the algorithm is called like down, sometimes you gotta give to get something.

FoxitReader -- PDF Reader

http://www.foxitsoftware.com/downloads/index.php

They even have one for Linux

Linux LPR print command

http://www.cups.org/documentation.php/options.html
Printing On Both Sides of the Paper

The -o sides=two-sided-short-edge and -o sides=two-sided-long-edge options will enable two-sided printing on the printer if the printer supports it. The -o sides=two-sided-short-edge option is suitable for landscape pages, while the -o sides=two-sided-long-edge option is suitable for portrait pages:

lp -o sides=two-sided-short-edge filename
lp -o sides=two-sided-long-edge filename
lpr -o sides=two-sided-long-edge filename

lpr -P MP560LAN -o InputSlot=Auto -o Resolution=600dpi -o PageSize=Letter

Origami Bambi Deer

Bambi deer
http://go.squidoo.com/?id=1120X507259&url=http%3A%2F%2Fwww.origamiseiten.de%2Fforcher%2Fbambi.pdf

PC Performance Measurement

http://en.wikibooks.org/wiki/Guide_to_Unix/Commands/System_Information
http://www.cyberciti.biz/tips/top-linux-monitoring-tools.html
http://www.computerhope.com/unix/uuname.htm
http://www.bioinfo-user.org.uk/dokuwiki/doku.php/notes/unixsysteminformation
CPU info:
$ uname -X
$ psrinfo -v
$ dmesg
$ mount -v
$ prtconf -v http://www.computing.net/answers/solaris/solaris-machine-specification/3343.html
$ prstat # equiv. of top

seed physiology

angiosperm plant life-cycle
http://8e.devbio.com/image.php?id=525
Figure 4 Life cycle of an angiosperm, represented here by a pea plant (genus Pisum). The sporophyte is the dominant generation, but multicellular male and female gametophytes are produced within the flowers of the sporophyte. Cells of the microsporangium within the anther undergo meiosis to produce microspores. Subsequent mitotic divisions are limited, but the end result is a multicellular pollen grain. The megasporangium is protected by two layers of integuments and the ovary wall. Within the megasporangium, meiosis yields four megaspores—three small and one large. Only the large megaspore survives to produce the embryo sac. Fertilization occurs when the pollen germinates and the pollen tube grows toward the embryo sac. The sporophyte generation may be maintained in a dormant state, protected by the seed coat.

mitosis of microspores (1n) to microgametophyte / pollen grain (1n) yields
- vegetative cell (forms the pollen tube) (1n)
- generative cell 2 sperm (1n) cells

3 sets of mitosis (no cytokinesis) of megaspores (the only one that persists) (1n) to megagametophyte (2^3 = 8 in total)
- 3 antipodal cells (1n)
- 2 polar nuclei in the centre (each 1n)
- 2 synergid cells (1n) in egg aparatus
- 1 egg cell (1n) in egg aparatus

double fertilization (happens when pollen tube carrying 2 sperm cells reaches the egg aparatus):
- one sperm fuses with egg cells forming 2n zygote forming 2n embryo
- one sperm migrates from the synergid cell and fuses with the 2 polar nuclei forming 3n endosperm

key processes during seed maturation so seeds maintain vigor and longevity?
1. accumulation of food reserves
2. hardening of the seed coat (testa) (for structural support, interacts with wind, pathogens, prevent of loss by leaching) and accumulation of antimicrobial compounds
3. mechanisms of seed dispersal - flyers
4. synthesis of protective compounds that allow the seed to withstand some degree of water loss and adverse environmental conditions eg. antioxidants, antimicrobial, resistance to pathogens
5. dormancy - can happen for many years, needs signals or conditions to break the dormancy (eg. conifers at moist and low temperatures) or smoke

parts of the seed
1. embryonic root (radicle) and shoot
2. food reserve tissues or organs (cotyledons - legumes, endosperm - cereals and megagametophyte - pines)
3. seed coat (testa)


timing of events: stages of seed development
1. histodifferentiation - cell division
2. seed expansion (maturation) - genes for synthesis of reserves (proteins, lipids, carbohydrates) are expressed, deposition (depositing) of reserves
3 . maturation (drying) / dessication - water loss, embryo becomes quiescent state (metabolically inactive)

in situ - in place

gram-negative has lost the cell wall so does not retain the purple dye.
Most pathogenic bacteria in humans are gram-negative organisms. Classically, six gram-positive genera are typically pathogenic in humans. Two of these, Streptococcus and Staphylococcus, are cocci (coconut-shaped bacteria).

nutrients like sugar (sucrose) are transported to seed from parent plant via phloem, for peas, they are stored in the pods then later remobilized to the germinating seeds

classification of seed storage proteins:
- albumins - soluble in water
- globulins (7s vicilins and 11s legumins) - soluble in salt
- prolamins - soluble in alcohol
- glutelins - soluble in diulte acid and alkaline

ch 2 - the language of first-order logic

http://en.wikipedia.org/wiki/First-order_logic#Alphabet

Abbreviations:
(a \supset b) for (not a or b)
safer to read as disjunction than as “if ... then ...”
(a three-vertical-lines b) for (((a \supset b)) or ((b \supset a)))

well-formed formulas[3] or wffs

A is a subset of B and conversely B is a superset of A
* B is a proper superset of A; this is written as B\supsetneq A.

ai - ch3 - solving problems by searching

search algorithms are judged on the basis of:
- completeness - find an answer if it exists
- optimality - find the best solution
- time complexity - # of nodes generated, BigO
- space complexity - # of nodes stored in memory

complexity depends on
- b - branching factor
- d - depth of the shallowest solution
- m - maximum depth of any path in the state space

Naive / blind search algorithms (b = branching factor, d = depth of solution, m = max depth of search tree, l = depth limit):
----------------------------------------------------------
- breadth-first search - can be memory intensive, time and space complexity O(b^d), complete and optimum
-- uniform-cost - breadth first but expands the lowest path cost first (greedy), same complexity as breadth-first
- depth-first search - not complete, not optimal, can take a long time, time complexity O(b^m), space complexity is good O(b*m) only -- only keeps node of a single path in memory
-- depth-limited search - has a limit l to the depth, but don't know limiting constant
- iterative deepening search - calls depth-limited with increasing limits, complete and optimal, same complexity as DFS
- bidirectional search, start from begininning and end until they meet halfway, O(b^(d/2))

informed search algorithms:
- best-first / greedy search
- A*
- local beam search - like breadth first search

consistent (triangle inequality): h(n) <= c(n, a, gn) + h(gn)
- n is starting node
- gn is the destination node, in this case, the goal node
- h(gn) = 0
- non-decreasing function
- c(n, a, gn) - cost of action a going from n to gn
- h(n) = cost from node n to goal

admissible - underestimating true (minimum) cost, so it never overestimates the cost

JDownloader

Rapidshare made easy
http://jdownloader.org/

OGM Rip

http://www.my-guides.net/en/content/view/129/26/

VOB to AVI

Use k9copy
http://ubuntu-virginia.ubuntuforums.org/showthread.php?t=1007151