Helicos BioSciences recognized its first product revenue during the last quarter of 2008, the firm said last week in a filing with the US Securities and Exchange Commission, in which it also said it expects to book its first instrument revenue later this year (see In Sequence 3/31/2009).
Over the last few months, Helicos has installed its single-molecule sequencer in at least two institutes. In addition, company researchers, speaking at scientific conferences, have been detailing how the tool is being used in a variety of applications, and reporting progress on new technical developments, including paired reads.
Helicos also revealed in its 10-K filing that as of the end of 2008, it had reduced its full-time headcount to 71 employees, from 114 at the end of 2007. Helicos said last December that it would reduce its workforce by 30 percent in order to cut operating costs, but declined to say at the time how many employees would remain with the company.
In its filing, Helicos said that during the fourth quarter of 2008 it recognized $36,000 in revenue related to the sale of proprietary reagents to an unnamed customer.
The company placed two instruments in 2008, one with Expression Analysis, which returned the instrument after several months, and another one with Stanford University. Helicos said in the filing that it expects to recognize revenue from this instrument — the company's first instrument revenue — sometime in 2009.
Since the beginning of 2009, the firm has placed Helicos Genetic Analysis systems at the Broad Institute and the Dana-Farber Cancer Institute (see In Sequence 2/17/2009).
At the JPMorgan Healthcare Conference in January, company officials also said that they had received an instrument order from an institute at the University of Maryland that they expected to deliver and install within a month. Helicos President Steve Lombardi told In Sequence this week that the order came from a private research institute, the St. Laurent Institute, which has an agreement with the University of Maryland, Baltimore, where the instrument would be shipped. He declined to comment on the status of this or other instrument shipments, citing the company's recent policy not to provide financial guidance.
Paul Morrison, director of molecular biology core facilities at Dana-Farber, told In Sequence by e-mail this week that his institute's Helicos instrument arrived in mid-February, and that installation and verification was completed by mid-March.
Since then, he and his colleagues have run 144 samples, submitted by nine laboratories, in three sequence runs, the third of which started this week. Applications include whole-genome sequencing of E. coli and C. elegans, ChIP-seq, and digital gene expression.
At recent scientific conferences, Helicos researchers presented results for these and other applications of the instrument.
Helicos Chief Scientific Officer Patrice Milos mentioned at the Cambridge Healthtech Institute Next-Generation Sequencing conference in San Diego last month that the company collaborated with a "major genome center" last September to sequence bacteria at 80- to 100-fold coverage in a single channel of its 25-channel flow cell.
Last October, company researchers sequenced the 100-megabase C. elegans genome at 27-fold coverage, using seven channels of the flow cell, and first presented the results at the Advances in Genome Biology and Technology conference in February.
[ pagebreak ]
And in a collaboration with the City of Hope, Helicos explored multiplexed sequencing of the p53 gene for SNP detection, combining five barcoded samples in a single channel, according to Milos.
Helicos is also collaborating with the Armed Forces DNA Identification Laboratory of the Armed Forces Institute of Pathology in Rockville, Md., on sequencing human mitochondrial DNA sequences from the remains of unidentified soldiers from the Korean War, a project presented in a poster at CHI conference by a senior research scientist in the laboratory.
As an example of a "counting application," Milos mentioned in her presentation a collaboration with Brad Bernstein at the Broad Institute where researchers used the HeliScope platform to analyze lysine-trimethylated histone H3 in a ChIP-Seq experiment.
In another collaboration, they assessed copy number variations in a human breast cancer cell line from the distribution of approximately 100 million aligned HeliScope reads along the genome. The researchers are now "beginning exploratory studies" of CNVs in formalin-fixed paraffin-embedded DNA samples, she added, with the hope that the Helicos system could be used in the future to analyze samples from tumor banks.
Helicos has also explored using its system for transcriptome analysis. In a digital gene-expression experiment, where a single read, or tag, is generated for each transcript, company researchers found that data from the same yeast sample, generated in two different channels of a flow cell, were highly reproducible.
The company has also analyzed samples from the MicroArray Quality Control study by DGE, and has compared DGE results from a frozen and a FFPE tumor sample.
For RNA-seq, Helicos has developed two sample prep methods — one that starts with fragmented mRNA, another that uses intact mRNA. This week, the company released an RNA-seq dataset on its HeliSphere Technology Center website, derived from human brain and liver tissues, where researchers generated 16 million aligned reads per channel for each sample.
The company said in a statement announcing the data release that it has provided its RNA-seq protocol to "a limited set of customers" and expects to commercialize it "later this year."
For the last few months, Milos said, the company has also been developing and testing a method for paired reads where users generate a first read, add unlabeled bases in a controlled manner, and then generate a second read.
The company is currently using paired reads to assemble bacterial genomes, but they could also be useful for detecting splice variants, Milos said. The method will become available to customers during the second half of the year, she added.
Paired-end reads, which require a different library preparation, are still in development, she said, and will initially focus on 1- to 2-kilobase inserts.