The Future of Biomolecules

Cees Dekker laboratory’s graphene nanopore technology for sequencing and protein analysis

thefutureofbio — Wed, 21 Sep 2011 19:42:45 +0000

Cees Dekker’s laboratory at Delft University of Technology was the first to measure DNA properties as it passed the molecule through a nanopore in a single atomic layer of the planar molecule graphene. The technology, if successfully applied to sequencing, would displace on market sequencing methods by enabling longer read lengths and obviating the need for fluorescent labels and associated instrumentation. Additionally, there are protein analysis applications that would build on a sequencing capability.

As described in a research publication by Schneider and co-workers, the lab developed a method to place single and multiple layers of graphene across micron-scale holes in silicon nitride, subsequently using electron beams to create 2 nm to 40 nm diameter holes in the graphene sheets. After placing the nanopore arrangements in microfluidic cells along with 16 micrometer-long double-stranded DNA and using a voltage to drive DNA through pores, the Dekker lab measured both the total translocation time of DNA through a pore and the time-dependent ion current. [1]

The researchers point out two applications: 1) sequencing at single-based resolution; and 2) detecting proteins bound to “biopolymers”[1]. This sequencing approach would compete directly with the development-stage ion-current nanopore sequencing technologies of Illumina/Oxford Nanopore and NABsys.

An evaluation of the Cees Dekker laboratory’s research on graphene nanopores is available on the Schiamberg Group website and includes: description of target applications, market context, and potential impact on one or more markets; analysis of technical challenges, applications, and likelihood of success; recommendations on actions that interested parties should take with respect to the laboratory; and recommendations on questions interested parties should be asking the laboratory. To purchase the Cees Dekker laboratory evaluation click here.

[1] Schneider et. al., “DNA Translocation through Graphene Nanopores,” Nano Lett., 10 (8), pp 3163–3167, 2010a (pre-print).

Halcyon Molecular’s electron microscopy for long-read sequencing

thefutureofbio — Wed, 07 Sep 2011 14:39:44 +0000

Start-up Halcyon Molecular is developing a method to sequence nucleic acids using high-atomic-number-labeled bases and electron microscopy. This approach to detection was first proposed by Richard Feynman around 1958. Halcyon is also developing a number of supporting techniques, including use of functionalized needles to stretch and place taut DNA on substrates for subsequent analysis. Schiamberg Group analysis and estimates indicate a high level of development risk for certain aspects of the technology and that Halcyon had approximately $18 million in cash on hand as of August 2011.

Halcyon is backed by venture capitalists and is attempting thus far to commercialize a sequencing instrument without commercial collaborations and corporate investment. Halcyon president Luke Nosek is a co-Founder of PayPal and Managing Partner at Founders Fund, which has portfolio companies including Facebook and SpaceX. The company plans to label nucleic acid strands with contrast agents that have a specificity or selectivity for certain bases, bind and stretch nucleic acids out onto a surface such that the bases are uniformly spaced at between 0.3 nm and 0.7 nm (preferably 0.5 nm), and then use electron microscopy to determine base sequence[1]. Halcyon is developing its technology for whole genome sequencing. It is targeting read lengths of 150,000 bases for human genomes[2].

An evaluation of Halcyon Molecular is available on the Schiamberg Group website and includes: analysis of technical challenges, applications, strategy, and likelihood of success; recommendations on actions that interested parties should take with respect to the company and questions interested parties should be asking the company; and data on capital raised, burn rate, cash available, and number of employees (based on our analysis and estimates). To purchase the Halcyon company evaluation click here.

[1] US 2010/0267157

[2] Halcyon company website’s post of excerpt from Genome Web article.

Edenspace Systems engineering plants to carry out own post-harvest degradation for biofuels production

thefutureofbio — Fri, 02 Sep 2011 14:55:09 +0000

Edenspace Systems is engineering plants to produce – internally – cellulose- and lignin-break down enzymes that would otherwise need to be generated and provided externally, at higher cost. The enzymes are to be localized within cellular compartments, to be released after plant biomass is mechanically ground. According to US patent 7696411 and patent filing WO 2009/155601, the technology covers application to plants and trees such as maize, wheat, switchgrass, poplar, and pine. The technology is from Mariam Sticklen’s laboratory at Michigan State University.

An evaluation of Edenspace Systems is available on the Schiamberg Group website and includes: description of target applications and potential impact on one or more markets; analysis of technical challenges, applications, strategy, and likelihood of success; data on capital raised, burn rate, capital available, and number of employees (based on our analysis and estimates in cases); and recommendations on actions that interested parties should take with respect to the company. To purchase the Edenspace Systems company evaluation click here.

Anellotech’s catalytic pyrolysis technology: Is it viable for biofuels production?

thefutureofbio — Wed, 16 Mar 2011 18:35:42 +0000

Start-up Anellotech is developing a catalytic pyrolysis technology to convert wood waste, sugarcane bagassee, and corn stover to aromatic hydrocarbons, to be used as raw materials for the chemicals industry and as fuel additives for gasoline producers[1, 2]. Anellotech has also indicated on its website that its technology can be applied to produce unspecified transportation fuels[1], and it has previously indicated that gasoline and diesel are possibilities[3].

The inherent complexity of Anellotech’s technology and the vast number of associated chemical reactions and purification steps strongly suggest that application of this technology to transportation fuel production is at least a decade from commercialization and carries serious technical risk. As a result, technical hurdles, including scale-up, are likely to hamstring Anellotech and other first generation developers of catalytic pyrolysis for biofuels production such as Khosla-backed start-up Kior. In this context, Anellotech’s plan to initially target the more tractable aromatic hydrocarbons is a good strategy, and may allow it to survive long enough to improve and apply the technology to gasoline or diesel production. Alternatively, it’s possible that Anellotech has already experienced technical challenges and low yields for gasoline or diesel, and is pursuing, by default, the only feasible near-term products.

Anellotech’s technology comes from the laboratory of George Huber at University of Massachusetts Amherst, and the aromatic hydrocarbons the company currently intends to produce include toluene, benzene, and xylenes[1]. Huber claims that the laboratory-scale process yields 50 gallons of product per metric tons of biomass, where the product is evidently a mix of these three aromatic hydrocarbons[2] (we are assuming this value is for dry tons of wood, given that Anellotech has used wood sawdust as feedstock in early tests[2]).

The laboratory-scale yield of 50 gallons per ton of aromatic hydrocarbons is unimpressive when used as a benchmark for the potential yield of a transportation fuel like gasoline using the Anellotech process. For Anellotech to make transportation fuel production commercially viable, major process improvements will be needed, as yields will only decrease as finished fuel products are generated and scale increases.

At the core of Anellotech’s technology is a chemical catalyst related to zeolite ZSM-5 and comprised of a porous structure[2]. According to US patent application 2009/0227823, which is listed on Anellotech’s website, useful materials include cellulose, hemicellulose, and lignin, as well as associated pyrolysis-derived components and their combinations[1, 4]. Although reference 2 indicates the catalyst structure is made up of silicon and aluminum, it’s clear from US2009/0227823 that, more precisely, the elements silicon and aluminum are present with oxygen, in the form of silica and alumina[4].

For additional information, note that Forisk Consulting and the Schiamberg Group have conducted a multi-client study to assess the wood-based liquid transportation fuel sector in the continental United States. This research evaluates projects based on technology risk, with profiles of firms, wood markets and technologies. For more about the study, “Transportation Fuels from Wood: Investment and Market Implications of Current Projects and Technologies,” or to purchase it, click here.

1. Anellotech company website.

2. Katherine Bourzac, “From Biomass to Chemicals in One Step,” Technology Review (2010).

3. “UMass Amherst, Anellotech pioneer conversion of bio-oil to chemical intermediates”, Renewable Chemicals Digest (2010).

4. US2009/0227823, “Catalytic Pyrolysis of Solid Biomass and Related Biofuels, Aromatic, and Olefin Compounds.”

Institute for Systems Biology proposes combining proteomics methods

thefutureofbio — Wed, 29 Sep 2010 18:55:55 +0000

Ruedi Aebersold and Hui Zhang of the Institute for Systems Biology outline a method in US 2010/0222233 to address two current limitations in protein analysis: the inability of microarrays to reliably quantify protein amount and the inability of mass spectrometry to detect proteins of low abundance. The ISB approach is essentially to use both methods in sequence, with a microarray step preceding a MALDI-MS step. The use of an initial array step to retain peptides of interest (including fragments of low abundance proteins) enables subsequent MS analysis of low abundance proteins, and detection settings can be tuned for specific protein fragments to further enhance sensitivity. As outlined in the patent, the challenge in quantification with microarrays is that target protein and array element pairs will each have different optimal binding conditions, and target proteins will have solubility variations that are also dependent on solution conditions. However, use of heavy isotope peptide standards of known concentration solves this problem: Loss of target peptide during washing steps on the microarray will be accompanied by similar loss of its heavy-isotope peptide-standard counterpart, but the exact starting quantities of the heavy-isotope peptides are known, so the ratios of resulting MS signals (heavy/light) will yield the quantities of target peptides.

The approach outlined in the patent is an advance to the state of the art, but remains limited as a proteomics tool since it only works for previously characterized proteins. Basically, you have to know what you’re looking for already. It’s worth pointing out that the idea of separating incoming molecules prior to mass spectrometry analysis is not new (e.g., GC-MS instruments were in widespread use beginning around 20 years ago). However, the protein chip’s high degree of specificity and high-throughput nature – when used in combination with mass spectrometry – is new.

Reveo’s direct DNA sequencing using tunneling microscopy

thefutureofbio — Wed, 25 Aug 2010 17:49:12 +0000

Reveo is developing an ambitious technology to stretch out and deposit taut DNA on conductive surfaces for electronic base detection using one or more STM tips and tunneling current measurements. The linearization and deposition of nucleic acid sequences will likely be done using molecular combing. Reveo’s approach requires atomically flat and positively charged substrate surfaces (e.g. self-assembled monolayers on gold substrates or treated graphite substrates). In addition to molecular combing, Reveo has proposed other methods to linearize DNA, including electrophoretic and hydrodynamic stretching and transfer printing[1]. Furthermore, Reveo has proposed to develop STM tips that are knife-edge shaped, where the smallest dimension is nanoscale[2, 3].

Schematic of Reveo's DNA sequencing method from US 2009/0121133

Much like with IBM/Roche’s proposed DNA transistor, Reveo’s competitive advantage is largely based on the potential cost reduction associated with avoiding labels and the possibility of exceptionally long read lengths. In principle, there’s not much difference between Reveo’s technology and that of IBM’s DNA transistor: both stretch and confine DNA to allow for tunneling current measurement of individual bases (albeit in different geometrical arrangements). IBM’s approach will likely be more reproducible and offer a higher degree of control over local DNA segment position. It’s possible that Reveo’s immobilization approach will reduce smaller-scale configurational rearrangement and motion compared to IBM’s approach, but Reveo’s method will likely be undermined by irregularities in DNA deposition. Reveo’s knife-edge tip design is intriguing, as it would avoid cumbersome issues of probe tip and DNA backbone alignment. Interestingly, even with nanoscale knife-edge tips it appears that individual, isolated DNA molecules would be required as a starting point for analysis, as simultaneous analysis of multiple strands would only be feasible if local DNA contour were uniform along an entire chain length (thus permitting the deconvolution of signals from multiple DNA strands given each individual strand would generation a periodic signal).

Outside of its entry into the Archon X Prize for Genomics, little is known about Reveo’s efforts in DNA sequencing. The company was founded in 1991, has spun out a number of companies, and holds over 300 patents across multiple technology and product areas[4]. Furthermore, although Reveo announced a partnership with the University of Washington (Babak Parviz’s lab) in 2006[5], a 2008 Nature Methods article describing Reveo’s technology did not reference the University of Washington[3]. The University of Washington was awarded a $1.5 million grant in 2006 to develop this technology and Reveo has cited this as its own funding [6, 2]. US patent application 2009/121133 described above lists Parviz as the inventor and University of Washington as the assignee; it’s quite possible that Reveo has priority rights to this and related, future University of Washington patents.

References:

[1] US 2009/121133

[2] http://www.reveo.com/node/309

[3] Blow, Nature Methods, vol. 5, p. 267 (2008)

[4] http://www.reveo.com

[5] http://www.reveo.com/node/88

[6] http://www.moore.org/grants-awarded.aspx

Mobious Biosystems’ single molecule sequencing

thefutureofbio — Wed, 14 Jul 2010 18:27:55 +0000

Mobious Biosystems was founded in 1999. It is developing instruments to detect single polymerase conformation or mass changes during the sequencing-by-synthesis process, using physical methods not dependent on the use of fluorophores. The start-up does not appear to have developed anything at this time that will significantly impact the third generation sequencing market. It does appear to have made progress on PCR and hybridization array technologies[1]. It is important, however, to note the impressive range of sequencing ideas and corresponding patents generated by founder Daniel Densham and the small company (see below). Conversely, it’s important to recognize that nearly all of Mobious’ technologies either currently are (or initially were) very early-stage and ambitious from a technical standpoint. Its set of proposed technologies span too many areas to be compatible with a start-up’s capabilities and resources. In fact, Mobious’ patent portfolio has the breadth one would expect from the likes of Roche Diagnostics or Life Technologies.

Schematic of SPR technology from US 2008/0014592

The following is a sampling of the core components of Mobious’ proposed sequencing methods and patents: 1) detect conformational changes in a single processing enzyme or a change in the polymerase’s mass (e.g. association with a nucleotide) using SPR, TIRM, or other light-based interrogation methods. In one embodiment, nucleotides are added sequentially, and in another, advanced blocking group chemistry is proposed to allow all nucleotide types to be present in the same reaction[2]; 2) measure polymerase dissociation rate from a target strand, leveraging differences in polymerase dissociation rate which are dependent on the presence or absence of a complementary base[3]; 3) detect conformational changes of an enzyme based on FRET[4]; 4) measure a single polymerase’s dielectric constant in order to detect conformational and/or energy level changes indicative of association with a specific nucleotide type[5]; 5) a variation on US2008/0014592 where a helicase is used to accomplish sequencing[6]; and 6) use of other advanced optical methods to detect enzyme conformational states (recent filing)[7].

At this time, Mobious should narrow its focus to only the most promising of its approaches and applications and open its business model to a variety of commercial partnership structures (if it has not done so already). Although fourth generation sequencing should not be ruled out, proteomics and other life sciences tools are likely a better bet. To access fourth generation sequencing, Mobious needs to be able to demonstrate the essential working components of a sequencing prototype at this time – and there should be clear competitive advantage versus other emerging sequencing technologies. Although many of Mobious’ proposed sequencing methods would be too expensive, time-consuming, or error-prone, there are a few strong approaches in the mix. Finally, it’s worth pointing out the degree of secrecy surrounding the company’s activities: there is a lack of information available on its current sequencing capabilities; and, although it lists University of Exeter Innovation Centre as its headquarters[1], the Centre’s tenant list does not appear to include Mobious for whatever reason[8].

References:

[1] http://www.mobious.com/

[2] US 2008/0014592

[3] US 7604963

[4] US 2005/0214849

[5] US 2008/0064035

[6] US 2008/0096206

[7] US 2009/0029383

[8] http://www.spaceforsuccess.co.uk/Tenants.html

J.P. Morgan Report on Next Gen Sequencing

thefutureofbio — Mon, 21 Jun 2010 18:46:17 +0000

J.P. Morgan’s recent equity research report on next gen sequencing (NGS) is based on its survey of 24 US and 6 European laboratory directors[1]. The survey focused on respondents’ needs and views relating to current and near-term products from Life Technologies (LIFE), Illumina (ILMN), and 454/Roche Diagnostics (ROG). Here are some of the important takeaways for those involved in third and fourth generation sequencing: the top three applications for NGS that respondents are “most excited about” are RNA sequencing, ChIP sequencing, and whole genome sequencing; the 30 labs surveyed expect to purchase or upgrade 136 NGS systems by the end of 2013; third generation systems will account for 47% of total NGS systems in these laboratories by the end of 2013; and, a significant portion of the microarray market will be taken over by NGS products. Although informative and carefully calibrated with respect to time frame, the report is at times confusing with respect to the topics of genome-wide association studies (GWAS) and microarrays. Specifically, the issue of replacement of microarray-based GWAS by NGS-based GWAS does not appear to be distinguished from the issue of a possible decline of GWAS as a research methodology altogether. In this author’s view, it cannot be overemphasized enough that there will be (and need to be) changes in the approaches to genome-based drug and biomarker discovery, so asking questions about if NGS will replace microarrays for specific applications may not be as important as asking questions like this: Will this application/approach be used anymore? Doubts about the utility of many genomics approaches are nothing new and have been gaining momentum recently (e.g., see a recent New York Times article on this subject[2]). Finally, one of the respondent’s responses about GWAS was particularly humorous and foreboding, “GWAS is in trouble. I suppose you could say the glass is 3% full.”

[1] J.P. Morgan, North American Equity Research, “Next Gen Sequencing Survey: What Laboratory Directors Are Saying About Next Generation Sequencing, GWAS and Stimulus” (2010).

[2] Nicholas Wade, “A Decade Later. Genetic Map Yields Few New Cures,” The New York Times (2010).

How to evaluate early-stage technologies: Part 2

thefutureofbio — Tue, 15 Jun 2010 15:15:16 +0000

What should companies, venture capitalists, and entrepreneurs be looking for when they search for useful technologies and evaluate a technology’s commercial potential? It’s important to answer these questions: Exactly how early stage is the technology? And, if successful, what part of an instrument, device, diagnostic, or therapeutic would result? Company executives and researchers often communicate using the terminology of products. In doing so, they may be conflating development-stage technologies with actual, functioning products. The motivating logic is circular and unsurprising: Investors and the market warn us repeatedly that we shouldn’t be fooled yet again by science that’s un-business-like. That it’s really products we’re all after, since products meet market needs and generate revenue. Almost all technology developers, scientists, and engineers know this, and respond: We have what you want, product we sell for revenue. Thus, it’s important to evaluate exactly where a technology resides on the development-maturity continuum, from idea to preliminary experiments to prototype to extensive/specialized testing to product (and the answer may vary across applications) – and to what extent “sales” and “revenue” result from sponsored-research collaborations or government grants. The rhetoric can sometimes go further, with start-ups implying that they plan to develop (or have developed) all of the components of a complex instrument or device (not just a specific part or reagent), or both the target and the drug. While it’s not impossible for a start-up to build something that is altogether their creation, it’s often unlikely and inadvisable. Furthermore, if the company owns less then they imply or claim, then future upside could be reduced due to licensing fees, fractional income from product sales, or incompatibility with other products and platforms.

LingVitae’s ‘genetic binary code’

thefutureofbio — Wed, 09 Jun 2010 16:50:17 +0000

Overview: Norwegian start-up LingVitae is developing a tool to translate biological data of interest into a form that can be more readily detected. It’s using a restriction and ligation enzyme system to cleave two end bases at a time from target DNA fragments of around 4- 40 bases[1], and effectively replace such two base combinations with predetermined, longer sequences of DNA (or DNA bound to labels), which are then concatenated into a new and much longer DNA strand. The process can be thought of as making a genetic binary code, where A, T, G, and C are replaced with, for example, 0-0, 0-1, 1-0, and 1-1 (where 0’s and 1’s correspond to distinct 10-base-long units). The resulting DNA concatemers would be amenable to hybridization with probe oligonucleotides[2]. In addition, a similar overall method could be used to convert protein sequences into nucleic acid sequences for detection[3].

Analysis: LingVitae’s technology is innovative and will no doubt find uses, though its applicability to third and fourth generation sequencing will be contingent on the success or failure of companies like Pacific Biosciences. Specifically, if third and fourth generation systems achieve a sufficient level of accuracy for single molecule detection by analyzing target molecules in their native state, then LingVitae’s technology will be less useful. LingVitae is also up against technical hurdles relating to the accuracy of its ligation-enzyme recognition and synthesis system. Furthermore, overall processing time (on the order of 24 hours for an entire eukaryotic genome[4]) is a downside which could severely limit use for whole genome sequencing. For genome sequencing, LingVitae’s technology appears to be a fit with platforms that are very high throughput but face detection accuracy problems. Thus, certain forms of nanopore sequencing would be a possible fit, given that detection methods such as ion current blockade are inadequate to resolve the identity of a single base within a DNA segment. The technology could therefore be a fit for Illumina (ILMN).

References:

[1] Blow, Nature Methods, vol. 5, p. 267 (2008)

[2] US20070254280

[3] US20090047744

[4] Company website: http://www.lingvitae.com/