I'm a senior concentrating in Biochemistry and Computer Science.
Biophysics 101 Assignments
101 Week 8
The Wall Street Journal article (Part I): The Most Expensive Drugs; Lucrative Niches: How Drugs for Rare Diseases Became Lifeline for Companies; Federal Law Gives Monopoly For Seven Years, Fueling Surge in Biotech Profits; A Teen's $360,000 Treatment
This article is a two-part series. It is extremely relevant - in fact to both personalized medicine and metabolic engineering. It makes it clear that the economic issues of the biotech industry and personalized medicine are non-trivial.
Summary Drugs that target diseases that affect less that 200,000 people are considered Orphan Drugs and in 1983 Congress passed the Orphan Drug Act of 1983 which gave companies that develop Orphan drugs a monopoly on the market for 7 years without having to file a patent application with the FDA and a 50% tax refund for R&D expenses, along with other protections from competition from generic drug makers. The result is that these markets have become very profitable for biotech companies. Drug prices have skyrocketed to as high as $600,000/year for drugs that must be administered for a lifetime. "Today, nearly half of all drugs produced by biotech companies are for orphan diseases." Amgen, Genentech, Genzyme were built on Orphan Drugs. Since the price of the drugs are prohibitive to most patients the cost is often passed on to insurance companies and corporate health plans (which then pass the cost along to their employees and customers). The high price of the drugs is a result of the protection the companies receive from competition, the relatively high cost of producing biotech drugs (this is where metabolic engineering comes in, as the advances in drug production have been in genetically modifying cells to [over]produce the specific drugs), and the difficulty in copying biotech drugs by generic drug makers.
Commentary The advent of personalized medicine will result in the further proliferation of small markets. In class, it has been mentioned many times that small markets aren't a profitable for drug makers - but this article says the exact opposite; the smallest markets have been the most profitable for the drug makers. However, with the increase in the number of smaller markets it will become increasingly unrealistic that the Orphan Drug Act of 1983 will continue to apply to drugs targeting these markets since this will result in burgeoning public expenditure on health care. On the other hand, an increase in public expenditure on health care may be more acceptable to the public if even more of the public belongs to a targeted market (since people will be more willing to cover the cost of "orphan" (personalized) drugs if there might be some that treat their diseases). Also, increased personal genomic capability will increase the number of markets drug companies can enter - i.e. increase the effectiveness of drug discovery research, and thus perhaps allow the public to lower health expenditure while at the same time maintain a high level of insentive for the biotech industry.
technologies: - microarrays, RT-PCR, reporter genes - orthology to other species (sequence data is becoming available)
problems: - misses tissue-specific functionality (only two cell-lines are being studied)
101 Week 7
getting multiple genomes, and what to do with them
Linkage Disequilibrium is the correlation of the occurrence of one SNP allele and the occurrence of another SNP allele.
Haplotype is a particular combination of these SNP alleles on a chromosome.
HapMap study: The International HapMap Consortium launched the International HapMap Project in October 2002 to document "common" human sequence variation; they looked at SNP alleles with greater than 5% frequency of occurrence.
results: Find 1 SNP every 5kb, except frequency of common SNPs is lower on X chromosome and on the non-recombining region of the Y chromosome and the mitochondrial chromosome (mtDNA). The Y chromosome and mtDNA are thus useful for population genetics. With no recombination, LD patterns (occurrences of SNPs on different alleles) are only related to mutations, and thus occur on different geneological branches. Thus, recombination rates are the significant determinant of LD and haplotypes. The human genome is composed of hotspots and coldspots of recombination.
101 Week 6
Pittman, J., et. al. Integrated modeling of clinical and gene expression information for personalized prediction of disease outcomes. PNAS, 101: 8431-8436.
This paper is interesting in that it provides an example of an attempt using genomic information (in combination with clinical data) to predict lymphoma and breast cancer patient survival.
They used expression patterns of "metagenes" ("aggregate patterns of variation of subsets of potentially related genes") to collect statistics on patient survival and then build markov trees based on this data.
101 Week 5
- started updating Project collab page with ideas relating to diagnostic tool ... one that could be updated and integrated with globally available data
101 Week 4
- Review and comments concerning Systems Biology, Proteomics, and the Future of Health Care: Toward Predictive, Preventative, and Personalized Medicine and ideas relating to it
also reviewed by Chiki
101 Week 3
Review of "Systems Biology, Proteomics, and the Future of Health Care: Toward Predictive, Preventive, and Personalized Medicine".
This article has a good introductory discussion of Systems Biology and the technologies that it utilizes and its applications to personalized medicine. The authors make a good point that is relevant to our potential project idea of thinking about software that can help people effect their lifestyles based on genomic data. They mention that environmental factors can also be monitored with developing technologies - just like we assume individual genome sequencing will be accessible to most people in the near future, technologies that will be able to tell people what kind of chemicals they are exposed to, what their diet is, etc. will be soon available. This can be integrated into our model.
The authors argue that systems biology will be crucial to elucidating the mechanism of disease, which is essential for prediction and prevention. They also argue that a systems approach can prove very useful in treatment as well since while current treatments are focused on single molecules (like inhibitors, etc.) "the behaviors of most biological systems, including those affected in cancer, cannot be attributed to a single molecule or pathway, rather they emerge as a result of interactions at multiple levels, and among many cellular components."
Biomarkers using blood plasma proteome patterns ("proteome signature") to distinguish normal organ function from pathogenic organ function Serum Proteome Pattern Diagnostics
Program using Artificial Intelligence pattern recognition was used to "learn" the healthy proteome Mass Spec distribution, and was able to accurately predict unhealthy distributions, thus diagnosing patients with ovarian cancer
signature has also been developed to distinguish patients with prostrate cancer from healthy patients
"Some researchers contest that SELDI is not sensitive enough, and captures only high-abundance proteins, and therefor is not suitable for measuring true cancer biomarkers"
Intro to Personalized Medicine
Intro to Personalized Medicine Good article summarizing the companies exploring this area: http://www.bio-itworld.com/archive/100902/path.html
A goal of Personalized Medicine is to figure out which variations in the human genome are responsible for various diseases/conditions. Figuring this out will enable the administration of drugs that are more appropriate for an individual genotype, diagnose diseases more accurately and efficiently, and screen patients for clinical trials, etc.
The problem is currently being explored by several companies (Millenium Pharmaceuticals, Inc., Cambridge, MA; DeCODE Genetics, Iceland; Sequenom, San Diego, CA; Genome Therapeutics, Waltham, MA; Variagenics Inc., Cambridge, MA; Compugen, Israel to name the few that are mentioned in the above article). The companies are comparing genomes of individuals and isolating genetic markers that distinguish diseased genomes from healthy ones. The most common markers are Single Nucleotide Polymorphisms (SNPs) and microsatelites. For example, genomes from young patients are compred to healthy, old patients and SNPs that distinguish the two are investigated for their relevance in disease. They use statistics and bioinformatics to isolate the genes that seem to correlate to specific diseases/conditions. The result is that genes involved in asthma, hypertension, schizophrenia, allergies, diabetes, osteoporosis, breast cancer, melanoma, etc. have been mapped.
It becomes clear that an important challenge is to categorize people/genomes into geneologic groups so that doing genetics and statistics is possible.
Also, to make sequencing more economically viable and efficient, it is important to be able to identify and sequence specific genes in patients, once we know which genes we are looking for.
The review Meyer, J.M. and Ginsburg, G.S. (2002).The path to personalized medicine. Curr Opin Chem Biol 6:434-438 talks about this in a little bit more detail.
This figure illustrates the major areas where personal genomics research is applicable: screening for patient predisposition to a disease/condition, screening for disease onset, diagnosing a disease, and treating and monitoring a disease by applying therapeutics catered more specifically to the individual.
The article also talks about recent research in Single Nucleotide Polymorphisms (SNPs) and the identification of haplotype blocks, which are linkages between SNPs the knowledge of which helps minimize the number of SNPs that need to be searched to establish effective biomarkers.
Some questions/comments: -How is a sufficiently representative sample of the healthy human population obtained? -A useful tool would be a way of measuring dynamic gene expression during disease progression in humans.